Neuropathic pain is pain that is caused by a lesion or disease of the somatosensory system. Typically disruption to the somatosensory system leads to loss of sensation, however in neuropathic pain there is paradoxical pain in the hypo-aesthetic area. As well as sometimes being difficult to diagnose, neuropathic pain is often severe, long-lasting, and resistant to treatment. 
Nociceptive pain is characterised by the noxious stimulation of Aδ mechanothermal and C fibres by algogenic substances. Neuropathic pain on the other hand is pain due to damage to the somatosensory system.
The IASP defines neuropathic pain as "pain that is caused by a lesion or disease of the somatosensory system". This definition deserves some discussion. Previously it was defined as "pain initiated or caused by a primary lesion or dysfunction in the nervous system." There are two key changes in the definition update. Dysfunction was replaced with disease, with disease referring to all types of abnormal conditions such as inflammation, autoimmune syndromes, and ion channel disorders. The term nervous system was replaced with somatosensory system in order to avoid confusion with other types of pain arising from the nervous system such as spasticity and rigidity of the muscles.
"Neurogenic pain" is an obsolete term. Neurogenic implied that the pain was generated in the peripheral nerve, however neuropathic pain is more complex than that.
Aetiology and Pathophysiology
Neuropathic pain includes multiple different conditions which differ in their aetiology and pathophysiology. In other words, different disorders of the somatosensory system may give rise to the same clinical picture but through different mechanisms.
There are two common ways of classifying neuropathic pain: by the underlying aetiology and by the affected level in the somatosensory system.
- The affected level can be anywhere from the brain down to the peripheral receptor. The most commonly affected sites are the peripheral nerves, nerve plexuses, dorsal roots, spinal cord, and thalamus.
- Causative conditions include metabolic diseases (e.g. diabetic neuropathy), infection (e.g. post-herpetic neuralgia), vascular disease (e.g. stroke), trauma (e.g. orofacial neuropathy), and cancer.
|Herpes zoster||Spinal cord injury||Multiple sclerosis|
|Plexopathies||Syringomyelia||Parkinson's disease (?)|
|Radiculopathies||Spinal stroke||Epilepsy (?)|
Another classification scheme is by pathophysiological mechanism, although many of these mechanisms are hypothetical. A single mechanism can be the cause of pain in different conditions, and also cause different symptoms; and furthermore different mechanisms may be involved in the same condition.
The pathophysiology involves a complex and redundant interplay of neural generators, circuits, and mediators. Neuroanatomical changes consist of both peripheral and central adaptive and maladaptive mechanisms. The adaptive processes include changes in pro-nociceptive and anti-nociceptive systems.
Neuropathic pain is variable between patients even in identical injuries and diseases. Around half of the overall variability is due to genetic predisposition.
Peripheral Nervous System
Nociceptors can be sensitised and silent nociceptors can be recruited. Increased afferent input can lead to sensitisation of the second order dorsal horn neurons. This "afferent barrage" can induce secondary sensitisation changes on second order neurons in the dorsal horn.
Neurogenic inflammation can play a role. In normal sensory experience, a noxious stimulus applied to nociceptive sensory endings is transmitted to the brain. Inflammation can lead to nociceptive endings becoming hypersensitive causing inflammatory pain. In this setting there is an increase in excitatory neurotrasmitters and neuropeptides such as histamine, bradykinin, serotonin, and glutamate. Peripheral nerve fibres when injured can be "sensitised" - i.e. they bring about a intense and prolonged ectopic afferent activity to the central nervous system.
Another mechanism is a severe loss of small fibre input such as in diabetic neuropathy or small fibre neuropathy. This can lead to a sprouting of myelinated fibres into the superficial "nociceptive" laminae of the dorsal horn.
The neuroma is one model of neuropathic pain. Neuromas form in sites where nerves have been severed, and microneuromas can form in injured nerves. Neuromas generate spontaneous afferent activity.
In PNS-triggered CNS changes, the terms neurectomy and deafferentation are frequently confused. Neurectomy refers to injury distal to the DRG, in which case the DRG survive for a long time and still generate impulses that can activate the CNS and evoke sensory experience. Deafferentation is when the injury is proximal to the DRG (e.g. dorsal rhizotomy, root avulsion, or ganglionectomy), central afferent terminal degenerate, and activity along the nerve or DRG cannot access the CNS. Both neurectomy and deafferentation can lead to neuropathic pain, with neurectomy due to both PNS and CNS changes, and with deafferentation exclusively due to CNS changes. Deafferentation can be seen in late stage postherpetic neuralgia and tabes dorsalis, otherwise it isn't an important process in most peripheral neuropathies. Sensory patterns may be different in deafferentation versus neurectomy. Experiments on dermatomes in monkeys in the 1970s, using remaining sensibility method (3 nerves severed caudal and 3 nerves severed cranial to the target nerve), found that lesions proximal to the DRG caused a larger anaesthetic area (remaining intact nerve had a smaller dermatome), while lesions distal to the DRG caused a smaller anaesthetic area (remaining intact nerve had a larger dermatome).
Central Nervous System.
Changes can be seen in the brain with altered central processing (altered modulation) and recruitment of areas that aren't involved in pain in normal states.
In the spinal cord there can be a reduction in the inhibitory neurotransmitter GABA, an upregulation of the opioid receptor inhibitor CCK, and a downregulation of GABA and opioid receptors. Sprouting can occur from the central terminals of damaged Aβ fibres from lamina IV to lamina II of the dorsal horn. Normally only nociceptors terminate in lamina II. Furthermore, there can be death of interneurons in lamina II.
Hyperalgesia can be brought on in normal individuals with blockade of large-diameter afferent fibres. It is thought that mechanoreceptive afferent input from the larger delta and beta fibres suppress the afferent input form small Aδ and C fibres and have an inhibitory effect on the dorsal horn. In neuropathic pain this inhibitory function may be impaired. This is hyperpathia, seen along with heightened reflex excitability. Both sensory experiences and reflex action occur in distressing exaggeration when peripheral tissues have access to the central nervous system only by the smaller fibres.
In some cases there is an increase of sympathetic activity with sympathetic neo-innervation of the dorsal root ganglia of the damaged afferents. This can lead to further sensitisation of nociceptors with sympathetic fibre activity directly stimulating activity in sensory fibres.
The prevalence of neuropathic pain has been estimated at 7-10%, but with marked variations between countries ranging from 1-18%. With the aging population it is expected that the prevalence of neuropathic pain will increase.
Partial or complete sensory deficit is a necessary part of neuropathic pain, but is insufficient to cause neuropathic pain on its own. There is a wide spectrum of sensory impairment seen with the changes in some patients only being detectable through quantitative sensory testing and not through standard bedside testing. All sensory functions may be impaired, however spinothalamic function loss appears to be required (cold, warmth, pinprick).
Ongoing pain (a positive symptom) can be present even with a completely insensate limb (a negative symptom), and this is termed anaesthesia dolorosa.
Evoked Pain: Allodynia and Hyperalgesia
- Main article: Allodynia and Alloknesis
The morphological and functional nervous system changes lead to abnormal sensory signs. Hyperalgesia and allodynia are classic findings of neuropathic pain. Hyperalgesia is the lowering of the pain threshold with an increase in response to noxious stimuli. Allodynia is pain elicited by a stimulus that does not normally activate the nociceptive system. Hyperalgesia reflects sensitisation of receptors, while allodynia is a central phenomenon mediated by large myelinated fibres.
There are three types of mechanical hyperalgesia and allodynia seen.
- Static allodynia and hyperalgesia: blunt gentle pressure on the skin elicits pain. Some term this static allodynia but others call it static hyperalgesia. This appears to be mediated by sensitised C nociceptors (and Aδ fibres?).
- Punctate hyperalgesia and allodynia: punctate hyperalgesia e.g. when tested with a stiff von Frey hair, and punctate allodynia e.g. when tested with a toothpick. This is mediated by sensitised Aδ fibre high-threshold mechanoreceptors.
- Dynamic allodynia: light brushing elicits pain. This is mediated by Aβ fibre low-threshold mechanoreceptors normally responsible for touch sensation, but in neuropathy can evoke Aβ pain due to central sensitisation.
There are two types of thermal hyperalgesia and allodynia:
- Cold hyperalgesia and allodynia: cold stimuli elicits pain. This is possibly due to a loss of cold Aδ fibres leading to cortical reorganisation
- Heat allodynia: warm and heat elicit pain. Some authors incorrectly call this heat hyperalgesia, but a lowering of the heat pain threshold means that pain occurs to a normally nonpainful warming. This process is possibly due to sensitisation of C fibres and their second order neurons.
Evoked Pain: Hyperpathia
The IASP define hyperpathia as "a painful syndrome characterised by an abnormally painful reaction to a stimulus, as well as an increased threshold."
Hyperpathia may occur with allodynia, hyperaesthesia, hyperalgesia, or dysaesthesia. Some view it has a type of hyperaesthesia, or even equate the two terms. There is an unusual variation in time and space. For time variations, gently tapping the back of the hand may feel dull as if the hand were in a thick glove, and this refers to an elevation of the detection threshold. But with repeated taping (e.g. two per second - 2Hz - for 10 to 20 seconds), temporal summation occurs i.e. "wind-up", and the sensation becomes stronger and stronger, until finally there is explosive pain which is due to central hyperexcitability. Variation in space refers to a spreading of the pain.
There are four main features of hyperpathia.
- Increased detection threshold: An increased threshold to noxious or non-noxious stimulation. This may be due to reduced afferent input.
- Delay: An abnormal latent period in the perception of the stimulus. This may be due to reduced large fibre afferent input.
- Summation: this refers to an increasingly painful sensation to a repetitive stimulus of steady intensity which unmasks hyperpathia. Summation can be seen as an explosive overshooting pain response that can occur with strong withdrawal movements and a vasomotor or vegetative reaction. There is a lack or insufficient relationship between the stimulus strength and sensation strength. There is poor localisation and the inability to identify the nature of the stimulus that elicited the pain, with a radiating sensation out from the point of sensation to wide adjacent areas. Summation may be due to afterdischarge in damaged sensory neurons, crossed afterdischarge in sensory neurons, coupling of adjacent nerves, and central sensitisation.
- Aftersensation: there is a long aftersensation or pain after the stimulus has ceased. This can occur for seconds, minutes, or hours after only brief periods of stimulation.
Hyperpathia is a clinical feature seen in the presence of axonal loss i.e. peripheral or central deafferentation. However, all of the causes of neuropathic pain can be associated with hyperpathia, and is also a feature of wind-up.
Evoked Pain: Temporal Summation and Aftersensations
Temporal summation is an abnormally increasing painful sensation in response to repetitive stimulation of steady intensity. It is the clinical equivalent of wind-up and is related to hyperpathia. With repetitive stimulation, pain threshold decreases, and the slope of the stimulus-pain response curve increases, thereby unmasking hyperpathia.
High frequency repeated nociceptive stimulation produces a temporal summation of the nociceptive afferent impulses originating from the C fibres due to their relatively slow conduction velocities. This results in an increase in the perception of second pain without lessening the sensation of first pain, which is conducted rapidly by Aδ fibres. The accumulation of nociceptive activity within secondary neurons of the spinal cord is called windup. Windup is distinct from spinal sensitisation. Wind-up is an increase in frequency of C-fibre discharge. It is a transient phenomenon but can produce spinal sensitisation and become persistent.
Wind-up pain or abnormal temporal summation can be elicited by a variety of stimuli such as mechanical, thermal, and electrical. It can also occur in many types of tissues. This finding is seen in many chronic pain states, not just neuropathic pain.
Aftersensation is thought to be related to wind-up pain. It refers to the persistence of pain long after the noxious stimuli has terminated. This is another clinical feature of neuropathic pain. Examples are the persistent burning sensation in post-herpetic neuralgia after light touch, and the exaggeration of pain following exercise.
Paroxysms and Spontaneous Pain
In some conditions the pain follows the distribution of the peripheral nerve, while in other conditions the pain may be more diffuse. Pain that is experienced in a numb site is called anaesthesia dolorosa and is highly suggestive of neuropathic pain.
Pain can occur spontaneously, i.e. be "stimulus-independent." This is often described as a constant burning. However, more typically pain occurs with stimulation i.e. is "stimulus-dependent", and this is termed "paroxysms" of pain. Paroxysms can be elicited by innocuous stimuli. Patients can experience shooting, electric shock-like, or stabbing pain.
There is no pathognomonic sensory descriptor for neuropathic pain. 50% of patients with musculoskeletal pain describe shooting pain and tingling sensations. Furthermore, 30% of patients with nociceptive pain describe burning pain.
In some cases like trigeminal neuralgia only non-noxious stimuli, not noxious stimuli, elicit the paroxysms.
There is often a significant difference between morning and evening in symptoms. The exact picture can depend on the aetiology.
Paraesthesiae can occur which are abnormal but nonpainful sensations. They can be spontaneous or evoked. Patients often describe them as pins and needles. It is thought that they reflect spontaneous activity of Aβ fibres.
Dysaesthesiae are abnormal, unpleasant, but not necessarily painful sensations. They can be spontaneuous or evoked. It is thought they dysaesthesia is mediated by sensitisation of C fibres.
Referred Pain and Spatial Summation
There can be an abnormal spread of pain. In painful myelopathic conditions a punctate stimulus can elicit a painful circular spreading sensation. There is a relationship between the area of spread and the intensity of pain. There is also a relationship between the intensity of deep pain and area of referred pain. Generally pain refers from deep to superficial structures.
This finding is thought to be mediated by changes in wide dynamic neurons in the dorsal horn. Wide dynamic neurons have small receptive zones that are excited by non-noxious stimuli. There is a surrounding much larger zone which responds to noxious stimuli. The larger receptive field zones overlap, and extend over several dermatomes. Therefore a noxious stimulus, but not a non-noxious one, activates several WDR neurons. Increasing the intensity of the stimulus results in further activation of WDR neurons in a cranial-caudal manner, and so there is progressive recruitment of WDR down the spinal cord. There may be a similar mechanism with sensory abnormalities in nerve injury, with spread to the contralateral side, as well as proximally and distally to the lesion.
There is often a significant negative impact on quality of life, daily function, and psychological wellbeing with anxiety and depression.
There are a variety of screening tools based on symptoms alone which can aid in distinguishing neuropathic from nociceptive pain. They include the Neuropathic Pain Questionnaire (NPQ), ID Pain, and PainDETECT. PainDETECT has only been validated in low back pain.
There are also tools that combine history and examination findings. They include the Leeds Assessment of Neuropathic Symptoms and Signs (LANSS), Douleur Neuropathique en 4 Questions (DN4), and the Standardized Evaluation of Pain (StEP). These tools have higher diagnostic validity than the above symptom only tools. StEP was validated in distinguishing nociceptive lumbar spinal pain from radicular pain, with a sensitivity of 92% and specificity of 97%.
Quantitative sensory testing (QST) is the assessment of response to external stimuli of controlled intensity. It can assess the function of large Aβ sensory fibres (through lemniscal ascending pathway), and small Aδ and C sensory fibres (through spinothalamic ascending pathway). It can differentiate between sensory gain and sensory loss. These are advantages over electrophysiological techniques (NCS and SEP) which can only assess Aβ fibres and cannot assess for positive phenomena.
There are many limitations of QST. Unlike electrophysiological tests, a response is required from the patient, and there is a clear influence of attention, fatigue, motivation, cognitive impairment. There is also the potential for malingering which requires null stimuli. It also can't localise the lesion along the neuroaxis unlike electrophysiological tests. The two assessment methods are complimentary. The IASP recommend avoiding QST in patients with language and cognitive difficulties, anxiety, or in cases of litigation and compensation as it is not possible to distinguish patients who are feigning symptoms from those who are not.
Detection and pain thresholds are assessed through applying stimuli with ascending and descending intensities. Quantitative sensory testing is very time consuming (it takes around 30 minutes) and requires expensive equipment and so is normally not suitable for clinical practice.
A comparison of QST and a quicker more cost-effective bedside version is outlined in the table below. The Faculty of Pain Medicine published a guideline of what they have termed "Pain Oriented Sensory Testing" They recommend only testing for dynamic mechanical allodynia (with cotton wool or a camel hairbrush), pressure-evoked mechanical allodynia (with fingertip pressure), punctate mechanical allodynia (with toothpick), and cold allodynia (with tuning fork). They also recommend testing for hyperpathia, but do not suggest routinely testing for hyperalgesia.
Zhu et al compared a cost-effective examination (slightly different to what the faculty of pain medicine recommend) to full QST and found a favourable but imperfect correlation.
When doing QST or the bedside version it is important to compare sides using the nonpainful side as a control, and also to compare sites. However remember that some patients may have neurological changes to their contralateral side but that is the best control we have.
|Fibres||Sensation||Finding Descriptor||Clinical Relevance||Testing Equipment and Instructions|
|Aβ||Touch||Dynamic mechanical allodynia||Common to most NP. Central sensitisation.||cotton wool or camel hairbrush - 2cm stroke over 1 second and repeat||Von Frey filaments||NCS, SEPs|
|Aβ||Vibration||Vibration detection threshold||Infrequent but strongly suggestive of NP||Tuning fork (128 Hz)||Vibrameter†||NCS, SEPs|
|Aδ||Pinprick, sharp pain||Punctate mechanical allodynia and hyperalgesia||Common to most NP. Central sensitisation.||Punctate mechanical allodynia - use a toothpick apply 2 stimuli per second (2 Hz) and repeat.
Punctate mechanical hyperalgesia - optionally tested with a neurotips needle.
|Weighted needles||LEPs, IENF|
|Aδ||Touch||Static or Pressure-evoked mechanical allodynia||Common to most NP, also observed in inflammatory pain. Central sensitisation.||Finger tip - apply pressure until blanching of nail bed for 1 second and repeat.||Von Frey filaments|
|Aδ||Cold||Cold allodynia||Infrequent but strongly suggestive of NP. Central sensitisation.||Stainless steel 128 Hz tuning fork prongs applied to the skin for 1 second and repeat||Thermode‡||None|
|Aδ||Touch, Pain||Temporal summation indicating hyperpathia||Central sensitisation, test for "wind-up"||Toothpick applied 2 stimuli per second (2Hz) for 30 seconds. Assess change in pre- and post- pain scores, and aftersensation.|
|C||Warmth||Warm allodynia||Infrequent but strongly suggestive of NP. Peripheral sensitisation.||Thermoroller or warmed C size battery at 45° applied for 1 second and repeat||Thermode‡||LEPs, IENF|
|† or other device providing graded vibratory stimuli
‡ or other device providing graded thermal stimuli
The diagnosis is made based on features of impaired neurological function. Cardinal is the loss of function such as numbness and this reflects direct nerve damage. There may also be exaggerated function which reflects loss of inhibition from nerve damage; this includes hyperalgesia, hyperpathia, hyperaesthesia, and allodynia.
In CRPS there is the disturbed sensory function seen in neuropathic pain, but also vasomotor, sudomotor, and trophic changes.
The diagnosis of neuropathic pain is reinforced by finding an aetiology for the pain such as diabetes mellitus or nerve injury. Electrodiagnostic studies may help in diagnosing certain conditions.
In some conditions the management is directed towards the underlying cause. This includes nerve entrapment syndromes and radicular pain. The response to treatment in these cases can be favourable.
Neuromas can be treated with variable success with resection, burying the neuroma, or ligating them.
In other aetiologies, the treatment options are often disappointing. The difference in aetiology and pathophysiology to nociceptive pain makes neuropathic pain particularly challenging to treat.
Generally an MDT approach is used and is based on targeting unhelpful cognitive patterns, providing coping strategies, and pain education. Physical therapy can be used to try and maximise function. From a biomedical approach there are a variety of interventions such as medication, nerve blocks, TENS, spinal cord stimulation, and others.
There is no medication that exists that has been proven to have long-term efficacy or tolerability for many cases of neuropathic pain. The tricyclic antidepressants and gabapentinoids are first-line agents in neuropathic pain. They are effective in postherpetic neuralgia and painful diabetic neuropathy. Pregabalin has been shown not to work for radicular pain. 
Topical lidocaine can be used in some cases.
- Wilson. Clinical Pain Management: Chronic Pain. 2008
- Neurogenic Pain. Encyclopedia of Pain 2013
- Marshall Devor. Peripheral Neuropathic Pain. Encyclopedia of Pain. 2013
- Kirk EJ, Denny-Brown D. Functional variation in dermatomes in the macaque monkey following dorsal root lesions. J Comp Neurol. 1970 Jul;139(3):307-20. doi: 10.1002/cne.901390304. PMID: 4317449.
- Denny-Brown D, Kirk EJ, Yanagisawa N. The tract of Lissauer in relation to sensory transmission in the dorsal horn of spinal cord in the macaque monkey. J Comp Neurol. 1973 Sep 15;151(2):175-200. doi: 10.1002/cne.901510206. PMID: 4355326.
- van Hecke et al, Neuropathic pain in the general population: A systematic review of epidemiological studies, Pain: April 2014 - Volume 155 - Issue 4 - p 654-662 doi: 10.1016/j.pain.2013.11.013
- John Scadding. Hyperpathia. Encyclopedia of Pain 2013.
- Faculty of Pain Medicine. Pain Oriented Sensory Testing. 2018
- Astrid Juhl Terkelsen and Troels Staehelin Jensen. Hyperpathia, Assessment. Encyclopedia of Pain. 2013
- Cruccu G, Truini A. Tools for assessing neuropathic pain. PLoS Med. 2009 Apr 7;6(4):e1000045. doi: 10.1371/journal.pmed.1000045. Epub 2009 Apr 7. PMID: 19360134; PMCID: PMC2661248.
- Scholz J, Mannion RJ, Hord DE, Griffin RS, Rawal B, Zheng H, Scoffings D, Phillips A, Guo J, Laing RJ, Abdi S, Decosterd I, Woolf CJ. A novel tool for the assessment of pain: validation in low back pain. PLoS Med. 2009 Apr 7;6(4):e1000047. doi: 10.1371/journal.pmed.1000047. Epub 2009 Apr 7. PMID: 19360087; PMCID: PMC2661253.
- Backonja MM, Attal N, Baron R, Bouhassira D, Drangholt M, Dyck PJ, Edwards RR, Freeman R, Gracely R, Haanpaa MH, Hansson P, Hatem SM, Krumova EK, Jensen TS, Maier C, Mick G, Rice AS, Rolke R, Treede RD, Serra J, Toelle T, Tugnoli V, Walk D, Walalce MS, Ware M, Yarnitsky D, Ziegler D. Value of quantitative sensory testing in neurological and pain disorders: NeuPSIG consensus. Pain. 2013 Sep;154(9):1807-1819. doi: 10.1016/j.pain.2013.05.047. Epub 2013 Jun 3. Erratum in: Pain. 2014 Jan;155(1):205. PMID: 23742795.
- Zhu GC, Böttger K, Slater H, Cook C, Farrell SF, Hailey L, Tampin B, Schmid AB. Concurrent validity of a low-cost and time-efficient clinical sensory test battery to evaluate somatosensory dysfunction. Eur J Pain. 2019 Nov;23(10):1826-1838. doi: 10.1002/ejp.1456. Epub 2019 Aug 28. PMID: 31325385; PMCID: PMC6852113.
- Cruccu & Truini. Tools for assessing neuropathic pain. PLoS medicine 2009. 6:e1000045. PMID: 19360134. DOI. Full Text.
- Zhu et al.. Concurrent validity of a low-cost and time-efficient clinical sensory test battery to evaluate somatosensory dysfunction. European journal of pain (London, England) 2019. 23:1826-1838. PMID: 31325385. DOI. Full Text.
- Backonja et al.. Value of quantitative sensory testing in neurological and pain disorders: NeuPSIG consensus. References 2013. 154:1807-1819. PMID: 23742795. DOI.
- Faculty of Pain Medicine. Pain Oriented Sensory Testing Guidelines. 2018.
- Enke et al.. Anticonvulsants in the treatment of low back pain and lumbar radicular pain: a systematic review and meta-analysis. CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne 2018. 190:E786-E793. PMID: 29970367. DOI. Full Text.