Nerve injury after peripheral nerve blockade—current understanding and guidelines

Learning objectives.

By reading this article, you should be able to:

• •Discuss the incidence of nerve injuries related to peripheral nerve blocks compared to the rate associated with various surgical procedures.
• •Explain the aetiology and mechanisms that contribute to nerve injury secondary to a peripheral nerve block.
• •Describe methods to reduce the risk of neurological complications associated with peripheral nerve blockade.
• •Outline the management of a patient with suspected neurological injury caused by peripheral nerve blockade.

Key points.

• •Mechanisms of peripheral nerve injury are often complex and multifactorial.
• •The reported risk for permanent or severe peripheral nerve injury with regional anaesthesia is extremely low.
• •Damage to the perineurium is likely to be the most important factor in the development of peripheral nerve block-related nerve injury.
• •Multimodal monitoring techniques along with careful patient selection may decrease the risk of nerve injury associated with peripheral nerve blockade.
• •Robust follow-up of regional anaesthesia patients is important for early detection of potential peripheral nerve injury and appropriate intervention. Society guidelines provide a structured approach to managing these patients.

Incidence of peripheral nerve injury

The precise incidence of peripheral nerve injury (PNI) is difficult to ascertain as a result of its rare occurrence and the methodological heterogeneity and quality of the studies available. For example, the definition of nerve injury, duration of follow-up, methodology (i.e. self-reporting compared with active surveillance, single or multiple institutions), and actual diagnosis of nerve injury vary between studies. The ASA Closed Claims Project provides a comprehensive collection of adverse events associated with regional anaesthesia (RA) in the USA, but unlike the UK National Audit Projects, the lack of a denominator prevents the calculation of accurate incidences of complications. Prospective multicentre RA registries are now addressing this problem.

Overall incidence of PNI after surgery

The risk for permanent or severe PNI after surgery appears to be extremely low. In one American tertiary centre, the overall incidence of PNI in more than 380,000 surgical operations was three in 10,000. PNI was associated with certain types of surgery, general anaesthesia, hypertension, and smoking but not peripheral nerve blocks (PNBs).¹ Other case series have demonstrated that PNI is nine times more likely to be a result of factors related to the surgery or the patient than related to a nerve block.²

RA-specific incidence of PNI

Unlike central neuraxial blockade, studies examining complications of PNB include fewer patients and mainly relate to upper rather than lower limb blocks.3, 4 Although it was previously suggested that proximal blocks were at higher risk for nerve injury (because of the higher proportion of nerve tissue compared with connective tissue), the most recent American Society of Regional Anaesthesia (ASRA) practice guidelines suggests there is insufficient evidence to fully support this theory.⁵

In the early postoperative period after PNB, patients may complain of persistent paresthesias. This is referred as postoperative neurological symptoms (PONS) and can be present in up to 15% of patients.⁶ Reassuringly, this rarely results in permanent injury with the incidence significantly reducing over time. The latest estimates of PNB-related temporary nerve symptoms suggest an incidence of 0–2.2% at 3 months, 0–0.8% at 6 months, and 0–0.2% at 1 yr.⁵ Providing patients with an exact figure for permanent PNI is difficult, and more robust data collection is required to determine the overall incidence. However, in reported studies, the range varies from 0.014% to 0.04%.⁵ As discussed below, this figure is likely to be higher in certain groups of patients.

Surgical procedure-specific incidence of PNI

Cardiac, general, orthopaedic, and neurosurgical procedures have all been found to be associated with increased risk of nerve injury (Table 1).¹

Table 1.

Incidence of neurological complications associated with different types of surgery

Neurosurgical	0.07%1
Cardiac	0.077%1
General	0.05%1
Orthopaedics5
Shoulder surgery
Arthroscopy  Shoulder arthroplasty	0.1–10% 0.8–4.3%
Elbow surgery
Arthroscopy  Replacement	1.7–4.2% (often transient) Up to ∼10%
Hip surgery
Total hip replacement	Approximately 1% (Transient injury to lateral femoral cutaneous nerve cut. 15–88%)
Arthroscopy	0.4–13.3%
Knee surgery
Anterior cruciate ligament repair  Total knee replacement	0.3–77% 0.3–9.5%

Anatomy of peripheral nerves

Knowledge of neural anatomy is vital to the understanding of nerve injury (Fig. 1).

Fig 1.

Anatomy of nerve

Peripheral nerves are structures consisting of numerous fascicles held together by a connective tissue layer, the epineurium. The perineurium is a multilayered epithelial sheath surrounding each fascicle. Inside each fascicle, the individual axons and capillary blood vessels are embedded in connective tissue, the endoneurium.⁷ The perineural cells have tight junctions and, along with the non-fenestrated capillaries within the endoneurium, create a protective barrier against chemical and mechanical injury akin to the blood–brain barrier. In contrast, the epineurium is a highly compliant, permeable structure. Interfascicular tissue embeds the fascicles and contains adipose tissue, blood vessels, and lymphatics. The vasa nervorum, or blood supply to peripheral nerves, consists of two independent, interconnected systems: the extrinsic supply consisting of arteries, arterioles, and veins which lie within the epineurium and the intrinsic supply consisting of the aforementioned non-fenestrated capillaries within the fascicles and endoneurium. The size and number of fascicles in a peripheral nerve vary substantially from one nerve to another, with the proportion of connective tissue which occupies the cross-sectional area of a nerve increasing as the nerve moves away from the neuraxis.

Intraneural compared with extraneural injection

Traditional teaching suggests that intraneural injection causes nerve injury. This was recently challenged by a controversial large case series of deliberate intraneural injections which resulted in no PNI, potentially leading to the misconception that intraneural injections are safe.8, 9 Crucially, the exact location of the needle tip is key in determining the likelihood of neurological injury. An injection that takes place outside the epineurium is considered extra- or perineural whereas any injection inside the epineurium is considered intraneural. An intraneural injection can be either extrafascicular (without breaching the perineurium) or intrafascicular (breaching the perineurium). This distinction is extremely important as damaging the perineurium exposes the protective environment of the fascicles.⁵ Even small amounts of solution injected intrafascicularly can lead to axonal degeneration and permanent neural damage.¹⁰

Fortunately, inadvertent intraneural injections often do not result in intrafascicular needle placement. A plausible explanation for this is needle deflection away from the tough perineurium, therefore avoiding fascicular injury. Extrafascicular intraneural injection is less likely to result in nerve injury.¹¹ However, society guidelines advise that the anaesthetist should not purposefully seek needle-to-nerve contact or intentional intraneural injection.⁵

The connective tissue that makes up the epineurium is distinct from the extraneural connective tissues surrounding the nerve. For example, the connective tissue surrounding the sciatic nerve forms a common extraneural layer referred to as the ‘paraneural sheath’. Injection deep to the paraneural sheath at the level of the sciatic nerve bifurcation around the tibial and peroneal nerves is considered extraneural as the epineurium is not breached and results in improved quality and speed of onset of block. Similarly with the brachial plexus, the extraneural connective tissue is contained within a sturdy fascia, the prevertebral fascia.

Pathophysiology of PNI

Severity of PNI

Severity of nerve injury is classified according to the degree of axonal disruption with the Seddon classification most commonly used. This defines nerve injury from mild to severe as neuropraxia, axonotmesis, and neurotmesis. Neuropraxia occurs when damage is limited to the myelin sheath and the rest of the nerve architecture remains intact. It is typically seen with excessive stretch or nerve compression and is associated with the best prognosis. Recovery is expected within weeks to months. Axonotmesis is the loss of axonal continuity but with an intact endoneurium. Recovery can be prolonged and incomplete, depending on extent of injury. Neurotmesis refers to complete transection of the nerve (axons, endoneurium, perineurium, and epineurium). This injury usually requires surgical intervention and is associated with the gravest prognosis.

Mechanisms

The traditional mechanisms of PNI related to PNB are described under four broad categories: mechanical, pressure, chemical, and vascular. Mechanical injury can result from partial or complete lacerations, forceful needle to nerve contact or intraneural injection. Intraneural injection can result in a combination of mechanical damage to axonal structures directly or the perineurium, which leads to a cascade of pathophysiological changes, high intraneural pressures secondary to local anaesthetic (LA) injection in a confined space, and LA-induced chemical neurotoxicity. Pressure is a relatively common mechanism of injury. Sustained high intraneural pressures can exceed capillary occlusion pressures leading to nerve ischaemia whereas chronic extrinsic compression of nerves by surrounding structures such as fibrous bands or scar tissue can lead to entrapment neuropathies. Vascular injury occurs because of direct trauma or acute occlusion of the arteries from which the vasa nervorum are derived. Inadvertent vascular puncture resulting in either an external or internal haematoma may also lead to ischaemia. Local anaesthetics and adjuncts can also reduce neural blood flow. Chemical injury is the result of toxicity secondary to direct injection of neurotoxic solutions (e.g. LA, alcohol, phenol) or an inflammatory reaction if injected adjacent to the nerve.

Aetiology of PNI

The aetiology of PNI is multifactorial and includes anaesthetic, surgical, and patient factors. Determining causation in the setting of surgery and RA can be challenging because these factors are often confounding. Some large studies have failed to link PNB as an independent risk factor for PNI, unlike both epidural and general anaesthesia.¹ Nevertheless, intrafascicular, high-pressure injections in particular are linked to PNI, and intentional damage and trauma to a peripheral nerve during PNB should clearly be avoided. The choice of local anaesthetic and additive may also be important. Surgical factors relate to improper patient positioning and other direct or indirect injury resulting in excessive traction, stretch, compression, ischaemia, contusion, or transection. Tourniquet may cause PNI by physical compression, ischaemia, or both. A postsurgical inflammatory neuropathy is also possible. The ‘double crush theory’ suggests that patients with preexisting neurological conditions are more susceptible to subsequent nerve injury from a secondary insult at a site remote to the original neurological dysfunction. Causes of preexisting neurological deficits are numerous and include entrapment, metabolic, toxic, hereditary, demyelination, or ischaemic neuropathies. Diabetic neuropathy is of particular concern because it may increase the risk of nerve damage by at least 10 times.⁵ Other factors such as anticoagulant therapy, peripheral vascular disease, smoking, hypertension, and vasculitis may also increase the risk of PNI.1, 12 Both ASRA and the Association of Anaesthetist of Great Britain and Ireland (AAGBI) have published recommendations regarding abnormalities of coagulation and RA.13, 14 Deep PNBs are deemed high risk, and society recommendations suggest that guidelines for neuraxial techniques be followed for these patients.¹³ A decision to perform a PNB in an anticoagulated patient should take into account the site, compressibility, vascularity, and consequences of bleeding, should it occur.

Reducing risk associated with PNB

Role of nerve localisation techniques

There are no human studies clearly demonstrating an advantage of one nerve localisation technique compared with another to reduce PNI. It has been suggested that a combination of nerve stimulation, ultrasound (US), and pressure monitoring, referred to as ‘triple monitoring’, may be of benefit, but this remains to be proven.¹⁵

Paraesthesia

Deliberately seeking paraesthesia was traditionally used as a method of nerve localisation. This is no longer standard practice. In awake patients, using paraesthesia as an indicator of nerve contact is unreliable. Absence of a paraesthesia does not exclude possible needle to nerve contact or prevent PNI, although paraesthesia at the time of PNB does increases the likelihood of transient neurologic symptoms after PNB.¹⁶ If paraesthesia is reported during a PNB, it should prompt immediate cessation of injection and needle repositioning.

Peripheral nerve stimulation

Various studies have demonstrated that peripheral nerve stimulation (PNS) does not predict needle tip to nerve location as accurately as once thought. For example, with the needle tip indenting the surface of a nerve under US guidance, no motor response was generated 25% of the time with a current of ≤0.5 mA.¹⁷ A motor response at a current of <0.2 mA is highly specific for intraneural needle placement but not sensitive because a current >1 mA can still be required to elicit a motor response after deliberate intraneural needle placement. If a motor response is obtained with a current of ≤0.2 mA, the needle should be repositioned. The major limitation with this is that nerve puncture and mechanical injury may have already occurred, but nonetheless it prevents any further pressure or chemical damage as a result of LA injection. Most authors suggest that nerve stimulation with a current intensity of between 0.2 and 0.5 mA with a pulse duration of 0.1 ms indicates a needle-to-nerve position that is sufficient for accurate and safe placement of LA.

Electrical impedance

This measures resistance to flow of an alternating current in a circuit. There is a clear difference in impedance between extraneural and intraneural compartments¹⁸; however, nerve puncture must again first occur to detect an impedance change. As there is no defined absolute impedance value for intraneural needle placement, this is not widely used at present.

Ultrasound

US has many advantages including visualisation of spread of injectate, reduced dose and volume of LA required, and decreased intravascular puncture and incidence of local anaesthetic systemic toxicity (LAST). US can detect nerve swelling, which should prompt cessation of injection, but it is unable to distinguish accurately between inter- and intrafascicular needle-tip location. This may account in part for the lack of evidence that US can influence the risk of long-term PNI.2, 19

Reliable visualisation of the needle tip is essential to prevent intraneural injection, and various technical developments and strategies to assist with this are described in Table 2.20, 21

Table 2.

Strategies to improve needle tip visualisation on ultrasound (US)²⁰

Needle design	Echogenic needle This uses textured needle surfaces to enhance visualisation, i.e. polymer coating, etchings, and notches on the needle surface
Needle beam angle	The angle at which the needle shaft and US beam intersect (with an in-plane technique) Optimal angle appears to be >55°
Needle bevel orientation	Visibility improved when bevel opening orientated to directly face US beam (0°) or to face 180° away from the beam
Needle diameter	Larger diameter (at expense of increased tissue and neural trauma and patient discomfort)
Hydrolocation/hydrodissection	A small volume of injectate is delivered which acts as a surrogate marker of needle tip position. This further aids in opening up the space between structures
US technology	•Electronic needle tracking • Compound imaging – Compound imaging involves acquiring multiple images of the same object and combining them into a single image • Electronic beam steering – The US beam is tilted relative to the transducer, increasing the needle-beam angle towards 90° • Colour Doppler detection of tip using the ColorMark™ device – Clips onto the needle shaft and produces vibrations which generate a signal assisting with visualisation

Injection pressure monitoring

Traditional subjective assessment of pressure using the ‘syringe-feel’ technique during PNB is highly inaccurate. Objective monitoring is possible using either commercially available devices or a technique of ‘compressed air injection’. Pressure monitoring is considered highly sensitive for intrafascicular intraneural injection but lacks specificity. A low opening injection pressure (<15 psi) indicates an extrafascicular or extraneural injection. Although a high opening injection pressure (≥15 psi) may be a consequence of intraneural intrafascicular placement of the needle, it may also occur with needle to nerve or tendon contact, needle obstruction, or tissue being compressed by a US transducer. An intraneural extrafascicular injection appears to be associated with minimal increase in pressure because of the low compliance of this compartment.²² To date, no case reports of clinically significant nerve injury have been reported with low opening injection pressures.

Equipment-related factors

Needle design

Bevel type and needle size are important determinants of nerve damage. Sharp long-bevelled (12–15° angle) needles are more likely to pierce the perineurium, whereas blunt short-bevelled (45° angle) needles may allow fascicles to slide away from the needle, potentially reducing neural injury. However, in animal studies, if short-bevelled needles penetrate the perineurium the extent of mechanical trauma far exceeds that produced by a long-bevelled needle. The degree of damage is also directly proportional to needle gauge, with larger gauge needles markedly increasing fascicular damage.²³

Local anaesthetics and adjuvants

All LAs are potentially neurotoxic in a concentration- and time-dependent manner. The primary determinant of whether neurotoxicity occurs is the site of application, with injury being greatest with intrafascicular injection. Neurotoxicity differs amongst different LAs. Studies suggest ester LAs to be more toxic than amide LAs, with ropivacaine having the lowest potential for neurotoxicity. Injury can also be caused by LA-mediated vasoconstriction (levobupivacaine > ropivacaine > lidocaine) causing ischaemia and further prolonging neural exposure to the toxic effects of LA.

Adjuvants such as adrenaline (epinephrine) are commonly used in PNB. At low concentrations (2.5 μg mL⁻¹), the β-adrenergic effects predominate resulting in an increase in neural blood flow. Adrenaline at concentrations >5 μg mL⁻¹ (one in 200,000) reduces blood flow in a dose-dependent manner.²⁴ When combined with LA, the negative effects on blood flow are cumulative, that is lidocaine 2% with adrenaline 5 μg mL⁻¹ will reduce neural blood flow by 80%. Other additives such as opioids, clonidine, dexamethasone, and neostigmine are not believed to be neurotoxic, but are nevertheless not licensed for perineural injection. Ketamine and midazolam appear to be neurotoxic at higher doses. Interestingly dexmedetomidine demonstrates neuroprotectivity in animal studies. None of the adjuvant agents listed are licensed for perineural injection.

Patient selection

Theoretical evidence suggests that patients with preexisting peripheral neuropathy may be at increased risk of nerve injury. Although preexisting neuropathy is not an absolute contraindication to RA, it should be considered when balancing the risks and benefits of a PNB. If a block is undertaken, the concentration and dose of LA should be reduced and epinephrine avoided.⁵ Using US to maintain distance from the nerve in order to minimise the risk is also of theoretical benefit.

Awake or asleep?

Although paraesthesia and/or pain on injection do not reliably indicate potential PNI, an account of such symptoms in a responsive patient is an important warning sign. Consensus guidelines recommend that nerve blocks in adults be undertaken in an awake, responsive patient, which allows for symptoms of LAST to be reported. However, exceptions are adults at risk of movement during the block procedure and children, where PNBs may be performed under anaesthesia.⁵

Diagnosis and management of patients with neurological injury

Early recognition of nerve injury and prompt risk stratification to prioritise those that require urgent attention (i.e. imaging and/or neurological consultation) are essential to afford patients the best chance of neurological recovery. Early surgical and/or medical consultations should be sought to allow identification of potentially reversible factors such as extrinsic or intrinsic compression (e.g. tight casts or dressing, compartment syndrome, haematoma, and occult perineural microhaematoma). Recognition of nerve injury may be more difficult where excessive sedation or pain limits examination, a continuous perineural catheter is present, or in ambulatory patients who are not followed up.

Diagnosis of suspected PNI is based on symptoms, history, and physical examination. Society guidelines provide a structured approach to the diagnosis and management of patients with a suspected nerve injury (Fig. 2).5, 25

Fig 2.

RA-UK algorithm for management of postoperative nerve injuries. Reproduced with permission from RA-UK.

Patient evaluation

Symptoms of nerve injury can present acutely when the block recedes or may only become apparent after several days or weeks. Symptoms vary depending on the origin and severity of the injury, and may range from mild intermittent tingling and numbness to more severe persistent, painful paraesthesia, and sensory loss, motor weakness, or both.

A thorough history and examination is vital to identify the injury, and determine single or multiple nerve involvement and motor and/or sensory impairment. This has prognostic implications as pure sensory deficits that occur within the territory of the peripheral block can be observed and often resolve within days to weeks whereas motor deficits are generally a more ominous sign. Examination may also out rule any signs of infection (temperature, swelling, erythema) especially if a catheter has been sited.

Diagnostic tests

Neurophysiological studies are tests to assess the function of nerves and muscles, and consist of electromyography (EMG), nerve conduction studies (NCS), or both. EMG studies examine the electrical activity generated within a muscle and are carried out to determine the muscle units affected by the injury. Nerve conduction studies assess the electrical conduction of impulses along a nerve. After stimulation of a nerve, a characteristic waveform of nerve conduction velocity is produced. NCS can localise the site of the conduction block and may confirm or refute that the PNI lesion is at the site of the PNB. It may not always be possible to distinguish between anaesthesia and surgery-related causes when the surgical incision site, block site, and tourniquet are in close proximity.

Electrophysiology studies can also offer prognostic information depending on the degree of injury found on examination (i.e. neuropraxia, axonotmesis, or neurotmesis).

Management

RA societies such as RA-UK and ASRA have published algorithms to assist clinicians faced with such a situation (Fig. 2). In summary, management of a patient with suspected nerve injury can be divided into two groups consisting of those with either: (i) mild or resolving symptoms (or persistent sensory deficits) or (ii) complete or progressive neurological deficits (and/or motor deficits).

If the physical examination reveals mild sensory deficits, reassurance is key to alleviate patient anxiety with follow-up within 4 weeks to review the need for further diagnostic tests. Early neurology consultation and radiological testing is advised if the examination reveals motor involvement, difficulty in localising or reconciling the injury with the expected distribution of the anaesthetic block or surgery, if there is complete absence of nerve function beyond the duration of PNB, any incomplete injuries with moderate or severe functional limitations or progressive neurological deficit. If symptoms fail to progressively improve, neurology consultation should also be sought.

Electrophysiological studies are typically delayed for 2–3 weeks as this allows for sufficient signs of Wallerian degeneration to appear. However, in consultation with a neurologist, these studies may be requested earlier as the presence of EMG changes in this ‘early’ period may signal preexisting neurology which is not attributable to an acute PNB-related event.

When incomplete or no improvement has taken place by 3–5 months, referral to a specialist peripheral nerve surgeon should be considered.

No pharmacological therapy has been demonstrated to enhance neuroregeneration, so treatment is limited to physical therapy to maintain muscle mass and prevent flexion contractures, along with analgesic therapy using neuropathic agents and non-narcotic analgesics.

Summary

Most postoperative nerve injuries are unrelated to RA. Fortunately, the vast majority of nerve injuries eventually resolve, with the risk of permanent injury being extremely low. The aetiology of PNB-related neurological complications is multifactorial; the underlying mechanisms are mechanical, vascular, pressure, and chemical-related. An understanding of the pathophysiology of nerve injury may allow clinicians to minimise adverse events. Damage to the perineurium is likely to be the most important factor in the development of PNB-related nerve injury. There is no evidence to support the use of one nerve localisation technique over another, but perhaps a multimodal, integrated approach is best, along with careful patient selection. The publication of guidelines by RA societies on the management of potential PNI provide a well-structured approach when faced with such situations in clinical practice.

Declaration of interest

The authors declare that they have no conflicts of interest.

MCQs

The associated MCQs (to support CME/CPD activity) will be accessible at www.bjaed.org/cme/home by subscribers to BJA Education.

Biographies

Doireann O'Flaherty FRCA EDRA is a post-CCST fellow in regional anaesthesia at University College London Hospital, whose clinical interests include regional and obstetric anaesthesia.

Colin McCartney PhD FRCA FRCPC is Head and Chair of Anesthesiology and Pain Medicine at The Ottawa Hospital and University of Ottawa in Ontario, Canada. He has practiced and published extensively in many areas of regional anaesthesia and acute pain medicine. He is an editor for the journal Regional Anesthesia and Pain Medicine and associate editor for the British Journal of Anaesthesia.

Su Cheen Ng DTM FCARCSI EDRA is a consultant anaesthesist and lead for the Block Room service at University College London Hospital.

Matrix codes, 1A01, 2G04, 3A09