Clinical manifestations

Presentation and course

There are several central pain syndromes with different clinical presentations and underlying neurologic abnormalities. Generally, central pain experienced by the patient is different from peripheral neuropathic pain. The onset of pain is usually delayed after the occurrence of the initial episode that results in damage to the central nervous system; onset of pain may occur during the phase of recovery from neurologic deficits.

Most patients have difficulty finding appropriate words to describe the unpleasant sensation; they indicate that the pain is different from any other pain they have experienced before. It may be described as “ice-burn” pain. The unpleasant abnormal sensations may be described as central dysesthesia, spontaneous or evoked. The evoked unpleasant sensation is triggered by movement, heat, cold, or touch, and may be described as allodynia. Examination of the sensory profile reveals various degrees of spinothalamic dysfunction, from a minimal degree detectable only by quantitative testing to gross dissociated sensory loss.

Central poststroke pain. Many of these symptoms of central pain have been described in patients with thalamic infarcts or the poststroke syndrome. The hemiplegia may improve during the early months, but spontaneous pain with overreaction dominates the syndrome and has its onset from 1 to 6 months after the stroke (average is 3 months). Rarely, the sensory symptoms may appear 1 year after the onset of stroke. Pain occurs on the hemiplegic side opposite to the lesion in the thalamus, but parts of the body may be spared. The hand is most frequently involved.

Central neuropathic pain after spinal cord injury. Spinothalamic dysfunction is the only common link between central pain due to cerebral and spinal lesions. Thermal pain threshold abnormalities are more pronounced in spinal cord pathologies than in cerebral lesions. Deficits of dorsal column function (touch and vibration) are common in central pain of supraspinal origin, but such deficits are rare in central pain due to spinal lesions. Central dysesthesia syndrome is seen in patients with incomplete spinal cord injury, but it is less marked than poststroke pain. In the anterior spinal artery syndrome, complete absence of temperature and thermal pain perception with preservation of dorsal column sensory function and a variable degree of motor function loss is usual. Pain in syringomyelia is usually experienced within the area of the dissociated sensory loss. There was no significant difference in the extent of sensory deficits between patients with or without neuropathic pain, indicating that lesions of the spinothalamic pathways do not adequately explain the development of central neuropathic pain.

Patients with spinal cord injury may present with mixed types of pain, such as musculoskeletal, visceral, and neuropathic. Most of the manifestations of neuropathic central pain are below the level of the lesion, whereas peripheral neuropathic pain with radicular distribution is more likely to occur at the level of injury.

Central pain in multiple sclerosis. This is quite common, and prevalence in various studies ranges from 40% to 80%. In addition to central pain, multiple sclerosis patients experience peripheral neuropathic pain such as dysesthesias in limbs, trigeminal neuralgia, and painful tonic spasms.

Central pain in Parkinson disease. Chronic pain is the most troublesome nonmotor symptom in Parkinson disease. It may be musculoskeletal due to motor disability and may appear at any time during the course of the disease. In 40% of the patients, the pain is neuropathic, both peripheral and central. It is described as stabbing, burning, scalding, or lancinating. It is unprovoked and may occur in unusual locations such as the face, genitalia, pelvis, or abdomen. Characteristics, severity, and distribution of pain varies with gender, race, as well as stage of Parkinson disease, and genetic polymorphisms might also play a role in pain severity (54).

Prognosis and complications

Development of central pain is a complication of injury to the central nervous system, but the predictors for the development of central pain are not definitely identified. It is difficult to identify stroke patients who are at risk of developing central neuropathic pain.

The prognosis of central neuropathic pain varies according to the primary pathology. In patients with severe neurologic deficits such as hemiplegia, neuropathic pain is usually not seen until the partial recovery from the neurologic deficit. Central pain may spontaneously diminish in intensity over the course of time; rarely, it may disappear completely after occurrence of further strokes.

The course of central pain is less likely to change in patients with spinal cord lesions, as there is much less recovery of neurologic deficits after injuries or vascular insults to the spinal cord than in cases of similar lesions in the brain. Results of a cross-sectional study with age-matched able-bodied controls indicate that the individual differences in the response to spinal cord injury and/or individual traits rather than the mere exposure to trauma may have a role in the emergence of central neuropathic pain and psychological distress and mood dysfunction (20). Results of a study to search for predictive biomarkers for development of central neuropathic pain after spinal cord injury in patients found that reduced pain inhibition capacity precedes and may lead to central neuropathic pain (19). Thus, at-level pain adaptation is an early central neuropathic pain biomarker to identify individuals at risk and initiate preemptive treatment.

Biological basis

Etiology and pathogenesis

Some of the neurologic disorders wherein central neuropathic pain occurs are listed in Table 1. These include a range of conditions such as stroke, tumors, injuries, and multiple sclerosis. Central pain is caused by a primary process in the central nervous system; the structure of the lesion is less important than its location. Selective spinothalamic injury caused by small lateral midbrain lesions is a very rare cause of central poststroke pain that can remain undiagnosed for years (08). Although spinothalamic dysfunction occurs in almost all patients with central pain, many poststroke patients with spinothalamic dysfunction do not develop central pain.

There is no consensus regarding the pathogenesis of neuropathic pain. A possible underlying mechanism may be dissociated sensory loss. Several new theories are based on the concept of deafferentation. Central pain may be due to abnormal hypersensitivity of damaged fibers, the generation of new receptors, alterations in the central pattern of impulses, and activation of secondary synaptic pathways.

Molecular biology of central pain remains largely unraveled. It has been postulated that noradrenergic and, to some extent, serotonergic central inhibitory systems are involved. This explains why amitriptyline, a noradrenaline reuptake inhibitor, is the most effective drug for the treatment of central pain. What is certain is that pathophysiology of central neuropathic pain is different from that of nociceptive pain, and it responds poorly to measures that are effective for nociceptive pain.

One explanation of central pain is that it is a disturbance of gamma-aminobutyric acid (GABA) and glutamate neurotransmission between the sensory thalamus and sensory cortical areas. A relative hypofunction of the GABAergic inhibition both at thalamic and cortical levels leads to an excitatory hypertonus in the same areas. This has therapeutic implications in that antagonizing the excitatory transmission or strengthening GABAergic inhibition may control central pain.

In multiple sclerosis, mostly patients who have lesions affecting the spinothalamocortical pathways run the risk of developing central pain (44). Anatomical studies have shown that the spinothalamic tract projects to the shell zone of the principal sensory nucleus and its rostral part. Physiological studies have revealed specific nociceptive neurons in some thalamic nuclei, from which fibers project to the insular and cingulate cortex. These structures may be subjected to excess excitation in patients with central pain. This hypothesis is supported by thalamic recording during stereotactic thalamotomy.

Under some circumstances damage to the peripheral nociceptive nerve endings may cause central pain. Molecular changes in the nociceptive terminals and ectopic firing of the afferent pain fibers at the level of the dorsal root ganglia may be the underlying mechanisms. Peripheral nerve lesions result in activation of various pathways leading to the production of inflammatory mediators in microglia or astrocytes in the spinal cord that sensitize dorsal horn neurons. Reactive glia in the cervical spinal cord following injury release proinflammatory substances that have been proposed as drivers of chronic central neuropathic pain (07). Dysfunctional glia or gliopathy is an important contributor to neuropathic pain following spinal cord injury (22). Microglia, resident immune cells in the CNS, initiate neuroinflammatory responses and play a role in the development and perpetuation of chronic central pain (28). A feedback mechanism is likely to enhance and prolong neuropathic pain even in the absence of ongoing peripheral stimulation. Injuries and infections may produce death or apoptosis of neurons in the nervous system; this in turn might induce neuronal sensitization and loss of inhibitory systems leading to neuropathic pain. Changes in voltage-sensitive Na+ channels may contribute to changes in nerve membrane excitability. Neuroimmune activation contributes to the development of chronic neuropathic pain in spinal cord injury. Concentration of tumor necrosis factor alpha, a proinflammatory cytokine, is elevated in the blood of these patients and could be a potential diagnostic biomarker of central neuropathic pain (60).

In conclusion, several parts of the nervous system involved in the physiologic generation of pain inputs, such as peripheral nociceptors, dorsal horns, spinothalamic tracts, thalamus, and cortex, undergo functional reorganization with altered nociceptive processing in association with chronic neuropathic pain. These locations provide the best targets for therapeutic intervention.

Epidemiology

Neurogenic pain is experienced by 1% of the population; no figures concerning the proportion with central pain are available. The incidence of central pain has been studied according to the primary disorder. The prevalence of central pain in stroke patients has been reported in 5% to 50% of cases. The frequency of poststroke shoulder pain is approximately 30% (02). The exact prevalence of chronic poststroke pain is not known, partly because of the difficulty in distinguishing this syndrome from other types of pain that can occur after stroke, such as shoulder pain and painful spasticity (32). In a prospective study of survivors in Helsinki Young Stroke Registry, late persistent central poststroke pain was found in 5.9% of patients and was associated with other types of pain (25). Although more than 50% of the patients with severe spinal cord injury have pain, less than one third of these have genuine central pain. Central neuropathic pain occurs in approximately 30% of patients with multiple sclerosis (45).

Prevention

Differential diagnosis

Confusing conditions

In peripheral neuropathic pain, an increase in ascending nociceptive pathways can be seen, whereas in central pain, a decrease in the activity of these pathways exists. Differentiation between peripheral and central forms of neuropathic pain is shown in Table 2. Both peripheral and central forms of neuropathic pain may coexist, making the differential diagnosis difficult.

Table 2. Differential Diagnosis of Peripheral and Central Forms of Neuropathic Pain

Diagnostic workup

Diagnosis is based on clinical presentation. The basic diagnostic workup varies according to the nature of the primary lesion, whether cerebral or spinal. The peripheral nervous system and sympathetic system should also be examined for any associated abnormalities. Neuropathic pain scale scores are based on patient responses to questions about pain intensity: 0 indicates no pain, 10 indicates the most pain imaginable. This is for use only in patients who have already been diagnosed with neuropathic pain. It is a comprehensive method for assessing neuropathic pain and may be particularly useful for assessing response to therapies.

Imaging studies. MRI is most useful in demonstrating lesions in the central nervous system of patients with central pain. In central poststroke pain of thalamic origin, lesion mapping using high-resolution MRI and a digital atlas of the thalamus is useful for identifying patients at risk of developing pain early after thalamic insults if there is a high odds ratio at the ventral posterior nucleus-pulvinar border zone, an area that is crucial in the pathogenesis of thalamic pain (53). PET may demonstrate the reduction of metabolic activity in the thalamus in patients with thalamic pain; however, peripheral neuropathic pain may also be associated with this finding.

Sensory examination. Components of the special examination regarding central pain are:

(1) Detailed clinical sensory testing for all modalities with particular emphasis on spinothalamic tract function.

(2) Quantitative sensory testing.

(3) Pharmacological tests; these tests are done to confirm the neurogenic nature of pain and predict the response to spinal cord or deep-brain stimulation procedures. These include the morphine test, Pentothal test, and transcutaneous electrical nerve stimulation test.

(4) Three-dimensional fiber tracking of the cervical spinal cord in patients with syringomyelia shows a direct relationship between central neuropathic pain and objective biomarkers of spinal cord damage; it is clinically relevant for the sensory assessment of such patients (27).

Management

Guidelines for the management of central neuropathic pain are listed in Table 3.

Table 3. Guidelines for the Management of Central Neuropathic Pain

Pharmacotherapy. Categories of first-line drugs for central pain include tricyclic antidepressants and selective serotonin-noradrenaline reuptake inhibitors. Opioids and anticonvulsants are also used. None of the available drugs can adequately control central neuropathic pain, which is often resistant to pharmacotherapy.

Amitriptyline. Tricyclic antidepressants remain the most effective drugs for treatment of central neuropathic pain. The efficacy of this drug in poststroke pain has been established in controlled clinical trials. Amitriptyline does not act by its antidepressant effect, but rather as a noradrenaline reuptake inhibitor. The usual starting dose of amitriptyline is 10 to 25 mg at night for a few days, and then increased to 25 to 50 mg. After a week, the dose can be increased to 50 or 75 mg, depending on the toleration by the patient. Effective pain relief may not occur for several weeks.

Selective serotonin reuptake inhibitors are less effective than tricyclic antidepressants due to their high selectivity. Venlafaxine is the most promising of this category of drugs because it has a better side-effects profile than amitriptyline, but further investigation is required to establish its role in management of neuropathic pain.

Opioids. Morphine and methadone have limited efficacy in treatment of central neuropathic pain. A systematic review of controlled clinical trials showed only very low quality evidence that oxycodone is of value in the treatment of painful diabetic neuropathy or postherpetic neuralgia (17). Supraspinal central neuropathic pain may respond less well to a trial of opioids than spinal central neuropathic pain. Titrating to higher doses of potent opioids may be considered before they are judged to be unsuccessful for refractory supraspinal central neuropathic pain (49).

Anticonvulsants. Various anticonvulsants have been used extensively in the treatment of central pain. These include carbamazepine, valproic acid, and gabapentin. Lamotrigine is included in the list of anticonvulsants approved for the treatment of central neuropathic pain, but a randomized, double-blind, placebo-controlled trial did not support the use of lamotrigine in patients with multiple sclerosis.

Second-line treatment involves the use of several agents, including the following:

Mexiletine. Mexiletine has been shown to be effective in poststroke central pain. It can be used to replace amitriptyline therapy (if ineffective) or can even be combined with it. If used as an outpatient treatment, a small starting dose of 50 mg twice daily is recommended, with weekly increments of 50 to 100 mg. Beneficial effect is usually seen at a dose of 400 to 600 mg/day. Patients need to be monitored, and electrocardiographic examinations are mandatory. One third of the patients discontinue therapy due to severe nausea and vomiting. Hypotension is another adverse reaction. Hospitalization is recommended if treatment is to be started with higher loading doses.

Third-line treatment involves the use of several agents, including the following:

Intravenous local anesthetics. Intravenous ketamine, a noncompetitive NMDA receptor antagonist, is primarily used for the induction and maintenance of general anesthesia, but it has also been used for the management of central neuropathic pain. Randomized controlled trials have shown that intravenous lidocaine relieves central pain. A possible mechanism of pain relief following ketamine treatment is regression of the NMDA receptor to resting state, which inhibits the propagation of nociceptive signals to the brain. A patient with Ehlers-Danlos syndrome and spinal cord ischemic myelopathy resulting in intractable central pain syndrome responded to a 1-week course of ketamine intravenous infusion under controlled generalized sedation in the intensive care unit, resulting in marked reduction in the use of analgesics including opioids (36). The evidence for efficacy of ketamine is moderate but in situations in which standard analgesic therapies have failed ketamine is a reasonable option. One drawback is that ketamine cannot be repeated too often because of its adverse effects.

Transdermal buprenorphine. Results of treatment of central neuropathic pain syndromes with buprenorphine are encouraging (21).

Transdermal capsaicin. Neuropathic pain after spinal cord injury is often refractory to standard treatments. Capsaicin 8% transdermal patch, an FDA-approved treatment of neuropathic pain in postherpetic neuralgia, has been shown to be effective for neuropathic pain in spinal cord injury in off-label use (55).

Clinical trials of treatments for central neuropathic pain. As of February 26, 2021, 97 clinical trials for “central neuropathic pain” are listed on ClinicalTrials.gov. Some of the publications relevant to these trials include the following:

Nonpharmacological measures. Positive relaxation (ie, by group therapy or use of audio tape) can be a useful adjunct to pharmacotherapy. Cognitive behavioral therapy has also been tried. Transcutaneous electrical nerve stimulation as an adjuvant therapy has been shown to be helpful in some cases. Acupuncture has also been used for relief of central pain, particularly following spinal cord injury (58). A Cochrane Database review of randomized controlled trials found insufficient evidence for the efficacy of nonpharmacological interventions for chronic neuropathic pain due to spinal cord injury (06).

Neurosurgical approaches. A historical review of neurosurgical attempts to control central pain describes several procedures that have been tried and abandoned (59). The first procedure to be performed for central pain, spinothalamic cordotomy, is no longer used because central neuropathic pain at the spinal level is a complication of this procedure. Resection of sensory cerebral cortex for thalamic pain has little to recommend from a physiological point of view because pain is interpreted at the thalamic level. There are reports of some relief of pain by frontal lobotomy, but resulting personality alterations and undesirable psychic changes now rule out this option. Stereotactic mesencephalotomy has been abandoned because of recurrence of pain in two thirds of patients after 6 months, along with the occurrence of dysesthesias as a postoperative complication. Neurosurgery is considered a treatment of last resort for chronic pain resistant to all other therapeutic measures. Procedures that are still performed for central pain include the following:

Spinal cord stimulation. Spinal cord stimulation involves the electrical stimulation of dorsal structures within the spinal cord and has been used frequently for the relief of chronic pain. It is often used for nonneuropathic pain such as that of angina pectoris. This procedure has been used successfully for relief of severe neuropathic pain in both lower limbs secondary to idiopathic acute transverse myelitis that had failed to respond to conventional pain therapies. Based on current evidence, spinal cord stimulation may represent a valuable treatment option, particularly for patients with chronic pain of predominantly neuropathic origin and topographical distribution involving the extremities. An open study has shown that spinal cord stimulation provides improved pain control in patients with intractable central poststroke pain (03).

An implantable neurostimulation system for dual channel therapy is designed to aid in the management of chronic intractable pain of the trunk or limbs. It has been shown to be effective in controlling neuropathic pain at the spinal level. The device has been approved for distribution by the United States Food and Drug Administration.

Motor cortex stimulation. Motor cortex stimulation is carried out by surgically implanted epidural electrodes for chronic motor cortex stimulation. Functional MRI is a useful guide for the operative targeting of selective motor cortex areas in neuropathic pain. Motor cortex stimulation has not been universally accepted because of controversy about the indications and variations in the efficacy reported in different series of cases. A study has reported that burst mode of motor cortex stimulation is more effective than tonic mode and should be considered when tonic stimulation is not sufficient in refractory neuropathic pain (50).

Deep-brain stimulation. Deep-brain stimulation is usually carried out after implantation of electrodes in medial and lateral ventroposterior thalamic nuclei and periaqueductal gray matter. In a series, good results were obtained with relief. Deep-brain stimulation has not provided satisfactory relief of thalamic pain and post spinal cord injury patients. Deep-brain stimulation is used more often in patients with Parkinson disease rather than for treatment of central pain. In a patient with central poststroke pain, permanent electrode placement following initial pain relief by stimulation of periventricular gray region and nucleus accumbens led to sustained relief reported at 1 year of follow-up (37). A long-term efficacy study of deep brain stimulation showed that benefit and efficacy varied by etiology, with relief in 70% of the cases with post-stroke pain (05). Review of the literature and long-term follow-up of a small series of patients suggests that stimulation of the posterior limb of the internal capsule is safer and more effective than motor cortex stimulation in treating patients with chronic neuropathic pain affecting the lower limb as it is a more targeted treatment for the subset of patients whose pain and spasticity are referred to the lower limbs (14). Deep brain stimulation to treat neuropathic pain refractory to pharmacotherapy has regulatory approval in the United Kingdom, but not in the United States.

Stereotactic thalamotomy. Stereotactic thalamotomy is performed predominantly for movement disorders and rarely for pain. For neuropathic or deafferentation pain, the preferred target is usually the sensory nucleus of the thalamus at a site corresponding somatotopically to the location of the pain. Thalamotomy of the principal sensory nucleus and its rostral part has been shown to ameliorate central poststroke pain.

Noninvasive brain stimulation. Techniques for noninvasive brain stimulation include transcranial direct-current stimulation (tDCS) and repetitive transcranial magnetic stimulation (rTMS). Application of rTMS over the motor cortex has been claimed to produce short-term relief of some types of central neuropathic pain. An open study of combination of rTMS over the motor cortex with motor imagery exercises and mirror therapy reported a significant decrease in pain at 48 hours, which was a good indicator of long-term response to this treatment (34).

Transcranial direct-current stimulation involves the application of a weak electric current to the cerebral cortex through the scalp, modifying neuron membrane excitability, which may lead to neuroplasticity following daily sessions. Use of tDCS for central pain is controversial. Results of a sham-controlled randomized clinical trial on patients with chronic neuropathic pain due to human T-lymphotropic virus type 1 infection showed no significant relief (51). A systematic review of the literature, including clinical trials, indicated significant effects of tDCS on central neuropathic pain as compared to the control group, but no definite conclusions about the efficacy of the neuromodulating effect of tDCS due to inconsistencies in methods and diversity of results obtained (09).

Noninvasive cortical stimulation is limited by the limited duration of the analgesic effects and the need to perform the procedure in hospital settings. A study has proved that internet-based, long-duration tDCS at home is safe and technically feasible for providing long-lasting relief in 50% of patients with central neuropathic pain (15).

A consensus by Latin American and Caribbean experts, based on a systematic review of the literature, recommends noninvasive central nervous system neuromodulation by tDCS and rTMS for the control of central neuropathic pain, as it has no severe adverse reactions, even though the analgesic effects are low to moderate (04).

Novel cellular and molecular approaches in development. Several cell and tissue transplant as well as gene therapy procedures are in development for chronic pain, but none of these are suitable for application in central pain. The implants secrete analgesic substances and are refined methods of drug delivery.

Cell therapy. Chronic delivery of antinociceptive molecules by means of cell grafts near the pain processing centers of the spinal cord has been investigated for the treatment of central neuropathic pain (30). Based on animal experimental studies, neural stem cell transplantation is a potential option for relieving neuropathic pain after spinal cord injury by reducing overexpression of P2X receptors (11).

Gene therapy. Gene therapy approaches include the use of viral vectors for delivery of genes for analgesic peptides. Although no specific gene therapeutic approach for central pain exists, analgesic therapy may be provided in a manner adequate to control pain. Viral vector-mediated gene transfer to dorsal root ganglia could be employed to treat central neuropathic pain after incomplete spinal cord injury (29). Reduction of neuropathic pain by intrathecal administration of an antiinflammatory interleukin-10 nonviral gene therapy formulation (XT-101-R) in a rat model of relapsing remitting experimental autoimmune encephalomyelitis supports gene therapy as a clinical strategy for the treatment of neuropathic pain associated with multiple sclerosis (18). Intrathecal injection of AAV vectors encoding combined serine histogranin (NMDA antagonist) and endomorphin1 (opioid) constructs into animal models of spinal cord injury can provide long term attenuation of neuropathic pain without significant adverse effects and has a clinical potential (31).

Special considerations

Pregnancy

No information is available to indicate if the course of central neuropathic pain is altered during pregnancy. Similarly, the presence of neuropathic pain does not affect the course of pregnancy.

Anesthesia

Central neuropathic pain is not a contraindication for general anesthesia. Spinal anesthesia by subarachnoid block has been reported to induce severe lightning pain in the lower limbs of patients with neuropathic pain of central origin, and this can be relieved effectively by intravenous thiopental with an approximate dose of 1 mg/kg, whereas intrathecal morphine hydrochlorides are ineffective (57). There have been no published reports of a similar observation since this one. Induction of anesthesia with propofol in patients with central neuropathic pain is associated with concurrent alterations of cortical and sensory thalamic activity as shown on intracranial EEG recordings from the ventroposterolateral nucleus of the thalamus and from the motor cortex (56).