Abstract Phantom limb pain is a common symptom experienced by over 90% of amputees. It’s defined as a painful sensation from a part of the body that no longer exists. There are a variety of methods for treating this neuropathic pain, but at the moment there is no specific treatment to tackle the pain completely. A mixture of medications and therapies has been proposed and trialed including drugs, surgical treatment and neuromodulation. Nonetheless, it is essential that a specific mechanism is targeted to in order to achieve the best therapeutic method.
From central stimulation to mirror therapy, the development of a wide range of treatments available today is due to central mechanisms of phantom limb pain that provide major grounds for research. Other treatments, aside from those related to the central mechanism, are also worth considering in order to improve our understanding. Results of these trials currently suggest that mirror therapy seems to be the most effective treatment based on central mechanisms, with the remaining therapies giving rise to a range of positive to negative outcomes. Introduction
Phantom limb pain first arose during the sixteenth century by a French military surgeon, Ambrose Pare (Weinstein, 1998).
He described this as pain being perceived from a part of the body which no longer exists, therefore belonging to neuropathic pain syndromes. The phantom limb is generally described to have a tingling sensation and a definite shape that resembles the limb pre amputation. Moreover, some claim to feel it move through space in the same way that the normal limb would have, for example, walking, sitting and stretched out (Melzack, 1973).
Almost all amputees would report these non painful sensations immediately after surgery (Nikolajsen et al, 2005).
Initially, the phantom limb feels normal causing the amputee to use the limb for its would be usual purposes such as reaching out for objects. However, eventually sometimes the limb begins to change shape resulting to a change in type of pain and sensation reflecting the extent of neurological damage this causes (Melzack, 1973).
The spectrum of pain from a phantom limb ranges from rare, short lasting painful shocks to a continuous, excruciating pain where the subject feels intense perception of the absent limb (Flor et al, 2006).
It seems to be more severe and vivid in the distal compartments of the limb sharing a variety of characteristics such as throbbing, stabbing, cramping or burning (Hill, 1999).
Blakeslee and Ramachandran have stated that some people misidentify a limb for representing another, for example, one patient described her phantom arm of being “6 inches too short” (Ramachandran et al, 1998).
Similarly. the onset of pain varies between individuals ranging from immediately to many years after the amputation (Nikolajsen et al, 2005).
Some experience the limb to be telescoped into the stump until only the foot or hand could be felt (Melzack, 1973).
These variants of the pain has lead to a broader definition where phantom limb pain is now also applied to describe pain in regions which have been completely denervated but not amputated (Portenoy et al, 1996).
These findings provided the root of extensive research encouraging scientists around the world to experiment and understand this condition. However, it still remains as a poorly understood and challenging to tackle pain syndrome.
Recently, a review in 2005 has approximated that there were about 1. 6 million individuals suffering from limb loss in the USA with this amount set to double by the year 2050 (Ziegler-Graham et al, 2008).
This emphasises the importance of trying to establish a certified treatment or therapy in order to improve the quality of lives of those suffering. Currently, scientists and researchers have proven that there are a variety of ways to approach phantom limb pain in terms of treatment and management with variable success rates.
How effective each method truly is depends on the patient and potentially the basis of the treatment. Surveys of amputees have shown that any form of therapy has been portrayed in a negative light due to the reduced possibility of it making any difference (Sherman et al, 1984).
As a result it’s essential to highlight how each treatment or therapy works in order to improve the prevention of this pain. With treatments such as ‘mirror box therapy,’ scientists are attempting to tackle phantom limb pain from every possible aspect and mechanism there is.
This leads to the branching of other approaches such as pharmacologic, anaesthetic, neurostimulatory, surgical, psychiatric and psychological (Portenoy et al, 1996).
Thus far, no treatment has been fully proven in terms of its effect at relieving phantom limb pain. Experiments and investigations have been carried out over the years to reveal which method is most efficient with most conclusions directing towards central mechanisms. However, is this the most effective approach? This essay attempts to address and explore the various techniques of managing phantom limb pain focused around central mechanisms.
Pain In order to fully appreciate the workings of each treatment and therapy, pain itself must be understood first. Up until the 1960s, pain was understood as an inevitable response to any tissue damage. However, in recent years, great advances have been made in our awareness of the mechanisms that underlie pain and its possible treatment (Loesar et al, 1999).
Pain has been defined by the International Association for the Study of Pain as ‘an unpleasant sensory and emotional experience which we primarily associate with tissue damage or describe in terms of such or both’ (Mersky et al, 1994).
This definition incorporates two important factors; first it reflects a sensory experience with the individual’s cognitive response, secondly, it suggests a relationship which is neither uniform nor constant between pain and tissue damage. Pain causes an individual to withdraw from damaging experiences and to avoid similar situations in the future. Most pain resolves promptly once the painful stimulus is removed and the body has healed. However, there are some cases where the pain persists despite the removal of the stimulus and apparent healing of the body.
Similarly, pain may also arise in absence of any stimulus, damage or disease (Niv et al, 2007).
This diversity of pain experiences accounts for its difficulty in achieving an explicit definition. Pain is not a single experience that can be specified in terms of a defined stimulus condition. Instead it can be agreed that pain, like vision and hearing, is a complex perceptual experience (Melzack, 1973).
As a result, pain may represent a multidimensional space comprising several sensory and affective inputs characterised by a combination of different qualities and unique events.
However, there are methods of classifying different types of pain. The two main branches are nociceptive and non nociceptive. Nociceptive includes somatic and visceral pain whereas non nociceptive refers to neuropathic and sympathetic. Nociception is the process where afferent activity produced in both central and peripheral nervous systems by stimuli are capable of causing tissue damage This is first detected by nociceptors which can distinguish mechanical, thermal and chemical changes above a certain threshold. Once this threshold is exceeded the nociceptors are stimulated to transmit a signal along the spinal cord to the brain.
This triggers a multitude of autonomic responses and may result in a subjective experience of pain (Patel, 2009).
In addition to these subtypes, it’s important to consider that other types do exist, such as, referred pain. Following from stimulation of the nociceptors, there are several pain pathways which lead to the brain to represent the type of pain triggered. These pain pathways are described as a sequence including different tracts leading to the brain. It begins with the neuronal fibres emanating from the peripheral receptors and synapsing in the dorsal horn of the spinal cord.
Here, several tracts carry the nociceptive signals to the brain. Examples of these tracts include: the spino-thalamic tract, the most prominent ascending pain pathway terminating in the thalamus, the spino-reticular tract which terminates in the reticular formation and the thalamus and the spino-hypothalamic tract which projects to the supra-spinal autonomic control centres. Phantom limb pain can be categorized as chronic neuropathic pain syndromes which involves pain in a limb that has been amputated, or which results from partial or whole deafferentation.
It’s suggested that the onset and severity of phantom limb pain may be associated with the nature of the amputation. For example, depending if the amputation was due to chronic purposes or traumatic purposes the outcome may vary, although so far evidence for this is insufficient to draw a credible conclusion (Houghton et al, 1994).
Moreover, phantom limb pain is more widespread when amputation occurs during adulthood as oppose to in children and congenital amputees. This suggests a psychological background where greater attention is given to the lost limb creating ‘pain memories’ which may contribute to the pain further (Katz et al, 1990).
In addition to phantom limb pain, other derivative and surrounding pains exist, including stump and pre amputation. Stump pain is a common symptom of early post amputation, however in the majority of patients it subsides with healing. It’s defined as a sensation in the existing body part next to the amputated limb or deafferentation line. Furthermore, phantom and stump pain are an interrelated phenomena where reports indicate a greater prevalence of phantom pain among amputees with coexistent stump pain compared with amputees without stump pain. (Flor et al, 2006).
For a long while, the dominant theory for phantom limb pain was the irritation and inflammation at the nerve endings of the amputation labeled neuromas. Consequently, surgeons would amputate the stump further with intentions of eliminating the pain. However, as expected the pain would return, if not more severely, with two phantoms present for the same limb as oppose to the one original. This discounted the theory of neuromas to be accountable for the pain and lead to an explanation that it may be related to the changes of the plasticity of the central nervous system (Flor et al, 2006).
This lead to further research into the central mechanisms of amputees in order to fully grasp the changes and harm created. Central Mechanisms Currently, the most common explanations surrounding phantom limb pain are focused around central mechanisms. As well as other explanations such as peripheral and behavioural, central seems to have a major influence on the outcome of phantom limb pain. Since many changes occur here, it’s essential to identify what occurs at each level precisely. This provides a basis for potential treatment and therapy, despite the success rate still remaining ambiguous.
Changes at the level of Spinal Cord Evidence surrounding changes at the spinal cord in amputees is weak, however experimental evidence based on animal models have shown that neuropathic injuries do involve spinal cord processes. At the spinal cord the amputated peripheral nerve form communications with the neurones in the receptive field into the dorsal horn via axonal sprouts. Some neurons that aren’t responsible for transmission of pain also sprout into the dorsal horn in an area involved in the transmission of nociceptive afferent inputs (Flor et al, 2006).
This is followed by a process called central sensitisation where there is an increase in nociceptor activity associated with inflammatory pain. This corresponds to an adaptation in how responsive the synaptic neurones from the dorsal horn are, eventually creating an expansion of the neuronal receptive field. As a result, hyper-excitability and increased neuronal activity also occurs. During the course of this process, the activity of NMDA receptors also increases. Transmitters like substance P, neurokinins and tachykinins mediate this process at the dorsal horn. A process called the ‘windup phenomenon’ then occurs where there’s an up regulation of NMDA receptors.
Consequently, the firing pattern of the central nociceptive neurones is changed. This may initiate less local inhibitory mechanisms between segments in the spinal cord, resulting to nocioceptive inputs at the supra spinal centres as well as spinal disinhibition. Deafferation is also observed at dorsal roots which are injured or damaged from the spinal cord. This may correspond with the effects of peripheral denervation centrally. This leads to spinal hyper-excitability. Amputees with damaged cervical nerve roots show signs of hyper excitability in dermatomes next to the denervated limb, suggesting the hyper excitability has diffused.
The combination of the changes at the spinal cord and the increased afferent input have been proposed to result in the progression of phantom limb pain. Changes at the level of the Brain When it comes to changes in the brain, the most common cause is cortical reorganisation; changes in the functional and structural somatosensory cortex subsequent to amputation. During reorganisation, the cortical areas representing the amputated extremity are taken over by areas in both the primary somatosensory and the motor cortex (Baron et al, 2010).
The process of cortical reorganisation has been studied in human and animal models following deafferentation and amputation where it partly demonstrates why the stimulation of afferent nociceptive neurones from the stump or its surrounding area produces a response in the missing limb. There is now evidence for considerable reorganisation of primary somatosensory, motor cortices and in subcortical structures following amputation (Grusser et al, 2001).
Studies have shown that the degree of cortical reorganisation has a direct link to the extent of pain and size of the differentiated region.
As a result, Ramachandran et al used these findings to propose that painful and non painful referred phantom sensations are parallel to perceptual reorganisational processes in the somatosensory cortex. This lead to the idea of topographical remapping, which is a relationship identified between areas of sensation from the mouth to stimulation sites in arm amputees (Halligan et al, 1951).
However, studies have shown that topographical remapping has only been found in a small percentage of amputees (Grusser et al, 2001) suggesting that phantom pain and referred sensation in the phantom are associated to differing central process.
Similarly, another study further demonstrates how referred phantom sensations may be less related to cortical changes in the brain. It illustrates how the greater the shift of the mouth representation into the area that previously represented the arm, the more distinct the phantom pain. Another proposed mechanism is stemmed from the ‘body schema’ theory, initially discovered by Head and Holmes in 1912. This is a concept in the brain where the entire body is thought of as a template; meaning any modification to the body, such as an amputation, would result to the sensation of a phantom limb.
To develop this explanation further, Melzack proposed a hypothses in 1989 of the ‘neuromatrix’ and ‘neurosignature’. In order to conceptualise the neuromatrix hypothesis, a system of neurones within a branch is visualized. This system incorporates a variety of inputs including somatosensory, limbic, visual and thalamorcortical components. This instigates complex interactions of neural networks in the brain. These interactions do not reside in a specific region of the central nervous system which contributes to the difficulty of understanding phantom limb pain (Portenoy et al, 1996).
This then evokes pain or other experiences. Neurosignature, however is a cause of the deprivation of various inputs from the limbs to the neuromatrix creating an abnormal neurosignature which results in the generation of phantom limb pain. Treatment based on Central Changes After understanding the basic central changes which occur during phantom limb pain, potential treatments can be developed and observed. Some treatments surrounding central mechanisms are perceived to be a lot more effective than others.
In order to a reach a full judgment, each treatment must be discussed and observed separately. Studies and trials represent how patients have responded to each therapy, identifying which method of management is most effective. As a result it’s beneficial to consider these case studies when comparing treatments. Mirror Therapy A common therapy used today is mirror therapy, first unveiled by Ramachandran and Rogers-Ramachandran in 1996. This allows a patient to feel the imaginary movement of the removed body part to behave as a normal limb via a mirror reflection (Ramachandran et al, 1996).
The reflection of the normal body is ‘superimposed’ on the perceived position of the phantom, helping to integrate and reorganise the mismatch between proprioception and visual feedback of the removed body. Therefore enhancing this treatment effect for phantom limb pain (Weeks et al, 2010).
Rizzolatti used the idea of a mirror neurone to explain the method of this therapy (Rizzolatti et al, 2006).
He discovered that a mirror neurone in the cortex fires when a person acts and when a person observes the same action performed by another.
Consequently by mimicking the behaviour, the subject will experience not only sensation, but also the similar emotion. To relate this to phantom limb pain, a patient will enable to experience the same sense of feeling and emotion of the normal body part by simply observing its mirror reflection. This is said to relieve pain by interfering with motor intention, proprioception and visual system. For example, out of five patients with painful phantom hand clenching, four achieved short term relief when a mirror allowed them to unclench the ‘hand’ (Dolin et al, 2004).
However, a person without phantom limb pain and no amputations won’t be able to feel these sensory experiences since they benefit from a non mirror neurone block preventing any influence, whereas a patient with an amputation does not have this non mirror neurone block allowing the mirror to have its influence (Ramachandran et al, 2008).
Furthermore, Ramachandran attempted to tackle the pain by addressing the phenomena ‘learned paralysis. ‘ He believed the phantom to be ‘paralysed’ and occupying the same position pre-amputation.
The mirror box therapy attempts to negate this ‘learned paralysis’ by unlearning the phantom paralysis with the aid of a reflection of the limb. This allows the patient to receive visual feedback of every message sent to the phantom limb. These explanations suggest the positive outcome of this therapy. A case study was conducted where 22 patients suffering from phantom limb pain were given mirror therapy for 15 minutes for a total of 4 weeks. They each recorded the number and duration of pain episodes throughout the trial, giving results where 100% reported a decrease in pain (Chan et al, 2007).
This itself illustrates the positive outcome of mirror therapy in phantom pain sufferers. Central Stimulation Techniques Stimulating the central nervous system can be indicative for managing phantom limb pain. However, since these techniques are deemed expensive and invasive, it’s usually dealt with as a last resort when other easier interventions have failed. This includes stimulation of the motor cortex which was initially a method of treatment for central post stoke pain. It was only until recently, when this technique has been applied to other neuropathic pain syndromes, including phantom limb pain.
This involves an implantation of the neuro stimulator directly into the subcutaneous tissue which stimulates the pre central gyrus using epidural surgical leads (Nguyen et al, 2011).
The stimulation of the motor cortex is usually an extradural procedure where stimulation of the cortex responding to the upper limb is more feasible than stimulation of the lower limbs. This continuous stimulation is aimed to relieve the pain even though overall success is minimal (Wester et al, 1987).
Spinal cord stimulation consists of low voltage stimulation of the spinal nerves to block the feeling of pain.
Electrodes are placed in the epidural spas adjacent to the spinal area which is presumed to be the source of pain. Initially, a temporary electrical stimulator is implanted and over a specific time period, patients are carefully monitored to observe any pain reduction. Patients who respond positively then proceed to receive a permanent implant. Pain is reduced due to the electric current interfering with the pain signal from the brain. Clinical results indicate beneficial outcomes of this treatment in patients suffering from phantom limb pain, both short term and long term.
Positive results were observed 52. 4% of 62 patients at the two year follow up, however this declined to 39% in the five year follow up (Viswanathan, 2010).
However, considering how difficult phantom limb pain is to manage, even the slightest benefit can have a positive impact. Having said this, experiments and studies surrounding spinal cord stimulation have concluded that this technique is least effective with patients suffering from phantom limb pain as oppose to those suffering from other neuropathic pain syndromes (Kumar et al, 1998).
Due to the uncertain outcome, it’s strongly recommended to trial the stimulation on a patient before complete insertion of an implantable pulse generator, avoiding any potential inconveniences. Asynchronous stimulation allows the separation of cortical zones which otherwise are reorganised during phantom limb pain. In this case asynchronous stimulation of the mouth which represents the invaded areas of the amputated arm and stump, separated the two cortical regions. As a result this reduced cortical reorganisation and therefore decreased phantom limb pain (Flor et al, 2000).
This provides promising future treatment methods. Other examples of central stimulation includes sensory thalamic stimulation, which in contrast to spinal cord stimulation, has the ability of blocking potential spontaneous neuronal activity. This has been shown to have the capacity to mediate phantom sensations with a reported 80% experiencing initial pain relief (Dettmers et al, 2001).
As a result, some propose this treatment to be more effective than spinal cord stimulation even though this has not yet been fully proven. A further development of central stimulation is deep brain stimulation.
The aim of deep brain stimulation is to identify where pain pathways aggregate in the brain to enable to destruction of associated structures which manifest the pain (Rokyta et al, 2012).
This procedure is considered invasive, has a high morbidity rate and low availability in comparison to other stimulation techniques. For these reasons, this routine isn’t recommended (Dolin et al, 2004).
Transcutaneous Electrical Nerve Stimulation (TENS) This method consists of a small device with two electrodes which are placed at the end of a stump to deliver electrical currents aimed to reduce pain.
The reasoning behind this may be that the electricity from the electrodes stimulates the nerves to send signals to the brain which disrupt normal pain signals. Another mechanism behind this, is that the electrical currents also stimulate the nerves to assist the body in producing natural painkilling endorphins. The electrical currents can be adjusted to different wave lengths according to the severity of the pain. However, with regards to phantom limb pain, the efficacy of this treatment is undecided. A study conducted by Mulvey et al in 2012 showed that the average pain intensity decreased on movement and at rest.
They recruited 10 individuals suffering from phantom limb pain to receive transcutaneous electrical nerve stimulation for 60 minutes over their transected afferent nerves. This study shows how TENS can potentially treat phantom limb pain however another study demonstrates how further trials are needed in order to conclude the effectiveness of transcutaneous nerve stimulation (Katz et al, 1991).
Destructive Lesions of Central Nervous System Dorsal root entry zone, abbreviated to DREZ, lesioning is another treatment linked to phantom limb pain consisting of surgery which destroys the area where damaged nerves join the central nervous system.
This intercepts pain messages from nerves to the brain therefore reducing any possible pain. A study by the Xuanwu Hospital in Beijing where 14 patients suffering from upper limb phantom pain were undertaken to evaluate their reaction from dorsal root entry zone lesions. Results of this study showed that 9 of the 14 (64. 3%) had satisfactory pain relief whilst the remaining 5 had poor pain relief experiencing no alterations in their phantom limb pain at all (Zheng et al, 2009).
These results aren’t sufficient to credit DREZ as an efficient method of managing phantom limb pain.
DREZ branches from cordotomy which has minimal literature supporting its use since it’s invasive and irreversible. Similarly, the development of later additional neuropathic problems can possibly arise (Dolin et al, 2004).
Other Treatments As mentioned earlier, it’s beneficial to consider other approaches of treatment aside from central mechanisms in order to appreciate the extent of techniques used and to provide a broader insight as to how this pain can be tackled. Alongside central mechanisms, the second most used approach would be peripheral.
Whether peripheral focused management is more effective than central is debatable but having an insight into this subject will better understanding. Currently, an array of treatments and therapies stem from the peripheral understanding of phantom limb pain. Examples include pharmacological drugs, and up and coming treatments like farabloc. Peripheral Mechanisms Over the years clinical observations have suggested that peripheral causes contribute to the phantom experience. Due to the strong positive correlation between stump and phantom limb pain, the events at the stump can be related to phantom pain.
For example, pressure on the stump can elicit the phantom experience. Similarly, infiltration of a stump neuroma can alleviate or abolish phantom limb pain. Amputation of the limb results in severing of the peripheral nerve axon with inevitable neuroma formation. These are said to produce abnormal spontaneous and evoked impulse activity (Dolin et al, 2004).
This ectopic, increased spontaneity is assumed to be a cause of an up regulation of sodium channels (Devor et al, 1993).
This coincides with changes in the dorsal root ganglion cells which also display increased sensitivity (Jensen et al, 1983).
Studies have provided evidence that in amputated limbs, the nerve endings remain sensitive to stimuli. Cooling these nerve endings increases firing rates and reduces blood flow causing its temperature to drop (Flor et al, 2000).
Patients who exhibit phantom sensations are said to display such patterns of limb temperature. This lead to the development of sympathetic blocks which increase blood flow to the stump reducing the overall intensity of the pain (Flor et al, 2000) Farabloc Farabloc is a woven mesh of stainless steel and nylon thread designed to shield from electromagnetic fields.
It’s essentially a metal sock made to wear at the end of an amputated stump in order to block any electromagnetic frequencies above 1MHz and allow any frequency below 1 MHz (Stannard et al, 2010).
The effect of this is to stabilise cellular permeability which in turn controls the release of inflammatory factors and modifies pain perception. Consequently, sufferers of phantom limb pain are able to wear this with comfort and experience pain reduction at the same time. Furthermore, this treatment can be applied to other neuropathic pain syndromes such as arthritic related pain and fibromyalgia.
The practicality of this treatment contributes to its efficiency allowing it to be more widely accepted by patients and healthcare professionals. Conclusion It’s apparent that these treatments share a positive impact on phantom limb pain with some evidently being better than others. Having considered the majority of possible treatments centred on central mechanisms, as well as an insight to peripheral, it’s now possible to reach an informed judgement as to which treatment has had the most effective outcome. The studies and trials mentioned provide a basis to make comparisons between different treatments allowing a more statistical measure.
Generally, treatments derived from central mechanisms range from being highly invasive, such as deep brain stimulation, to not being invasive at all such as mirror therapy. This demonstrates how judging the effectiveness of a specific therapy requires certain qualities be taken into account. These include; ease of availability, efficiency, duration of therapeutic effect, and overall effectiveness at reducing pain. Central stimulation techniques generally have a positive influence on managing the phantom experience, however there are several obstacles related to this technique.
For example, deep brain stimulation is very invasive with unpredictable outcomes, therefore putting the patient at risk during its course. As a result, the potential risks influence the cost of these procedures causing it to be one of the most expensive treatment options available. This already indicates how this solution isn’t efficient or widely available. Moreover, the variable success rates of all central stimulation techniques further impedes its effectiveness representing a decreased overall benefit. However, there are a handful of central techniques which are more welcomed than others.
For example, sensory thalamic stimulation is considered more effective than spinal cord stimulation with 80% of patients experiencing pain relief (Dettmers et al, 2001).
In addition, asynchronous stimulation and transcutaneous electrical nerve stimulation are considered almost equally as successful at reducing phantom limb pain, both being more cost effective and efficient than other methods. On the other side of the spectrum, mirror therapy has had a major influence on how phantom limb is approached and managed.
Ramachandran discovered how the pain wasn’t a consequence of neuromas being irritated by muscle activity surrounding the stump, but due to a further involvement of central factors. He concluded how in addition to peripheral inputs, there are central factors and other contributions to the pain. Further research into the cortical behaviour of phantom limb patients lead him to develop mirror therapy. Studies showing how 100% of subjects experienced some scale of pain relief successfully illustrates how beneficial this therapy is.
Alongside its high success rates, it also benefits from minimal equipment and preparation requirements further enhancing its efficiency and availability. However, although most results demonstrate positive outcomes, an extensive review by Rothgangel et al, 2011, has identified that the current studies aren’t sufficient to draw firm conclusions from since enough isn’t understood as to how and when mirror therapy should be applied. Albeit, it’s clear the results of mirror therapy are significantly greater than all other treatments discussed.