Phantom limb pain
Phantom limb pain (PLP) is frequently described by post-amputation patients as sharp, shooting, stabbing, burning, or cramp-like. Patients may also experience residual limb or ‘stump’ pain which too can become chronic. Stump pain is strongly associated with PLP, and, although treatment for each pain may overlap, it is important to differentiate between the two.
1 Recognize the risk factors for the development of PLP
2 Identify both pharmacological and non-pharmacological strategies for the long-term management of PLP.
2E01; 2E03
A 58-year-old gentleman with type II diabetes is listed for theatre for a below-knee amputation. He has diabetic nephropathy, background retinopathy, and peripheral neuropathy with reduced sensation of both feet. He has had rest pain at night for 9 months, secondary to his peripheral vascular disease. He describes a constant burning pain of the soles of his feet, despite them being cold to touch, and a throbbing ache in his calves worsened by walking.
He also has a non-healing ulcer on his right heel for 3 months which has been treated with antibiotics in the community. In this area, he experiences intermittent throbbing and a sharp, shooting pain radiating from his heel to mid calf. The area of the ulcer itself is numb, and he has reduced sensation to touch, distal to his ankle.
Other significant past medical history includes hypertension, hypercholesterolaemia, and mild COPD. He is a current smoker, with minimal alcohol intake. He had an uneventful general anaesthetic 13 years ago after injuring his arm in a DIY-related incident. He is overweight with a BMI of 32. His airway assessment is unremarkable, and he has a full top denture. His current medications include ramipril, simvastatin, insulin, aspirin, inhalers, and co-codamol 30/500 (two tablets four times daily), and tramadol 50 mg (as required).
Blood results show an Hb of 15.2 g/dL, WCC 6.9 × 109/L, Plt 353 × 109/L, Na+ 139 mmol/L, K+ 4.8 mmol/L, urea 8.1 mmol/L, Cr 140 micromoles/L, and an eGFR of 54. Coagulation and LFTs are normal. ECG shows a sinus rhythm, with borderline left ventricular hypertrophy.
He is concerned about PLP and would like to know more about it.
Phantom sensations are feelings perceived to be coming from the missing body part, such as itching, tingling, or movement, and can often be painful. Phantom pain has been reported in many different amputated body parts as well as limbs, such as breast, penis, rectum, bladder, eyes, and teeth. Up to 80% of amputee patients experience PLP, usually occurring shortly after the operation. Some patients have resolution of symptoms after a few months, but, in many, the symptoms persist. Occasionally, patients have no phantom symptoms, until many years later when it may spontaneously arise or be triggered by an event such as trauma (physical or emotional) or even a spinal anaesthetic.
PLP appears to be more prevalent in patients who had pain in the limb prior to amputation. Indeed, the pain is often similar in nature. It is also more common in lower limb and bilateral amputees.
The exact cause of PLP has been deliberated and investigated for many years, and unsurprisingly, like most pain states, it appears to be multifactorial.
The brain and nervous system display neuroplasticity and are constantly processing information, adapting, and restructuring as a consequence. This shapes our individual pain experience and is influenced by many factors such as past experience of pain, genetics, the environment, social and cultural background, and personality. This is known as ‘the biopsychosocial model of pain’.
In patients with PLP, structural changes have been shown in the somatosensory cortex, dorsal horn of the spinal cord (central sensitization), thalamus, brainstem, and the peripheral nerves. These changes are usually characterized by an increased excitability of neurones, combined with a reduction of inhibitory pathways.
PLP (again, like many pain states) can also be mediated by the sympathetic nervous system, causing an exacerbation of pain in response to emotion and stress. Furthermore, in the residual limb, neuromas develop, secondary to nerve trauma (surgery), and can also be triggered spontaneously by an increase in circulating catecholamines.
Another common finding is changes in the somatosensory cortex where the area representing the missing limb is ‘remapped’ and receives sensory messages from other areas. This can result in patients feeling phantom sensation and phantom pain when a different area of the body is stimulated.
A lot of focus on preventing PLP has been around the perioperative period and whether we can prevent or modify the changes that take place at the time of amputation. Several strategies have been studied and mainly involve minimizing perioperative pain signalling. High levels of acute pain have been associated with the development of chronic post-surgical pain, as have nerve damage and pre-existing pain—all factors common to many amputees.
The evidence is often conflicting, as frequently the studies have been of relatively small patient numbers or suffer from inadequate randomization. Follow-up data are often difficult to interpret, due to the patient population characteristics. Patients who commonly present for limb amputation have significant comorbidities, and, in many studies, the patients do not survive to the endpoints.
Epidurals and spinals undoubtedly provide superior acute pain control for amputation, but it is unclear whether they can influence the development of PLP. An ongoing epidural local anaesthetic (up to 3 days) may have a beneficial effect on long-term outcomes, as may a combination of epidural clonidine and diamorphine. Epidural ketamine does not appear to have an effect on PLP.
Most studies looking at this strategy were examining a catheter placed perineurally, with an infusion for up to 72 hours post-operatively. Unfortunately, no studies have shown a long-term reduction in PLP with this method. In addition, analgesia for acute stump pain was found to be inferior to an epidural local anaesthetic. This would be a useful adjunct to systemic analgesia for those patients unsuitable for central neural blockade.
IV ketamine and oral gabapentin have also been tried for the prevention of PLP, with disappointing results. Most patients in this group have pre-existing pain, and it is likely that blocking pain signalling at the stage of surgery does not reverse the existing central changes. It may be possible to prevent further changes secondary to nerve damage, but evidence for perioperative prevention is limited.
Your patient underwent his amputation under spinal anaesthesia with diamorphine plus a femoral nerve block with a catheter, followed by a local anaesthetic infusion for 2 days. He was very comfortable throughout and was able to participate with physiotherapy the next day.
Nine months later, you meet this patient again at his first appointment at the chronic pain clinic. He has adapted well to life with an amputated limb and is mobilizing with a prosthesis for short periods. His post-operative course was uneventful, and he is considering returning to work soon. However, he has troubling pain where his right leg used to be, particularly at night. His sleep quality is poor, and he finds himself getting up to sleep in a chair on most nights. He feels exhausted and unable to concentrate. He sometimes gets the pain during the day, particularly at times when he ‘really does not need it’, often at times of stress. His daytime pain is distracting, although this is less severe than at night. He describes the feeling as tense or gripping. His GP started amitriptyline at night which was titrated to 30 mg, but he was intolerant of the side effects such as dry mouth, urinary retention, and excessive sleepiness. He has started gabapentin, now at a dose of 300 mg three times day. Initially, he slept better at night but is unsure if it is having any real effect on his pain.
He has already described the characteristics of his pain, its variation during the day, the effect on his sleep, and the effects of analgesic medications. To complete a pain history, you ask him more about his pain. Is there anything he can do to make it better or worse? Does he have any other sensations? Does he have stump pain? Is his pain similar to his preoperative pain?
He reports that his leg feels better when he takes his prosthesis off and especially when he is relaxing in the evening. He does not think anything else changes his pain. His stump is tender after wearing the prosthesis, and occasionally he feels a tight sensation when at rest. He does admit that sometimes his pain feels like his preoperative burning pain but thinks this must be foolish, as the ulcerated foot has now gone.
It is important to elicit a detailed medication history. What other medication does he take and at what dose? Ask about each analgesic he takes, if it helps, if he has side effects. Are there any other medications he has tried? Has he tried other treatments like physiotherapy or acupuncture? His current medications are ramipril, simvastatin, insulin, aspirin, inhalers, and co-codamol 30/500 as before. He no longer takes tramadol, as it was ineffective, and he started gabapentin 2 months ago. His physiotherapy has formally finished, but he continues to follow their advice.
How does he manage with his daily activities? Find out what he can and cannot do. Who lives at home with him? Does he require any help or social support? What does his work involve, and how does he think he will manage?
Your patient is able to do most things around the house. He manages all personal care and is happy with transfers. He has a wheelchair for mobilizing and rarely uses it indoors now. He uses a stick for walking. He is at home with his wife who is relatively fit, and she does the housework and shopping. They also have a son and a daughter who are nearby and regularly visit. He owns a hardware store which he runs with his son who has been looking after the business since his father’s operation. He is keen to return to work and has been recently involved with the accounts side of the business. He does worry about lifting and moving stock and also about what he would do if he had a lot of pain when he was on his own in the shop.
This patient appears to be a stoic gentleman and initially denies any low mood. He states that ‘anyone in this situation would get a bit down’, but he tries to keep a positive outlook and wants to find a way to ‘deal with it’. After speaking to you at the pre-assessment visit, he expected he would have some phantom pain and has since read about it on the internet. He is concerned that it will not get any better and that he will continue to be too exhausted to go back to work. He wants to have good-quality sleep.
Many health professionals who deal with patients and their pain utilize validated questionnaires. These can provide useful information and can be a guide to success of treatment. Many of these tools exist; some examples are the Brief Pain Inventory, the McGill Pain Questionnaire, the Hospital Anxiety and Depression Scale, the Pain Catastrophising Scale, and the self-report version of the Leeds Assessment of Neuropathic Symptoms and Signs.
Examination is also important, and, for most pain patients, it will focus on the appearance and temperature of the skin, the sensation to light touch and pin prick, and, if applicable, the range of movement. Often a neurological examination and/or palpation of the painful area are performed.
Several studies have looked at the persistence of PLP and found that symptoms reduce in most patients. Around 60% of patients with PLP symptoms will have their symptoms remaining after 2 years, but the severity and frequency of the attacks will usually decrease. The pain is usually intermittent and is often similar in nature to the pre-amputation pain. Over time, the limb may feel shortened, a phenomenon known as ‘telescoping’.
There is little evidence for the effectiveness of simple analgesics, such as paracetamol and NSAIDs, in PLP. Antidepressants and anticonvulsants are effective for neuropathic pain, and gabapentin has some evidence of short-term benefit in PLP, but there is inadequate evidence for other medications in this class. Opioids have some effect on neuropathic pain, and studies have shown possible benefit for PLP. Short-term effects have also been shown in controlled studies of ketamine and calcitonin, although the numbers of patients were small.
Whilst there is little evidence for an effective pharmacology treatment, physical treatments have been shown to make significant improvements to a patient’s pain and quality of life. Many types of physical treatment, such as TENS, sensory discrimination, acupuncture, massage, mirror therapy, and graded motor imagery, have been utilized, with good effects, although evidence is limited. These methods often target the changes within the nervous system, attempting to manipulate the neuroplasticity of the brain and alter the cortical reorganization that has occurred in these patients. It is also important to establish any problems with the residual limb and prosthesis fitting which can aggravate the phantom limb.
Repeated surgery is not of benefit, unless there is an obvious pathology of the residual stump. Attempts in the past to remove neuromas have unfortunately worsened symptoms. Spinal cord stimulation has been used for refractory pain, with some success, but further studies are needed.
As for all patients living with chronic pain, learning new ways of managing life and coping with chronic pain can make significant differences to their quality of life. There are many resources available to provide an understanding of pain and ways of dealing with it. The biopsychosocial approach involves physiotherapy, psychology, and medicine which can be accessed individually or as part of the chronic pain service.
Pain specialist physiotherapy is about learning movements and techniques that increase function and ability such as pacing and goal setting.
Pain specialist psychology addresses stress, mood, thoughts, and feelings and teaches relaxation and techniques such as mindfulness.
Social support is also important, and there are many useful organizations and resources. There are local (e.g. the Pain Association Scotland), national (e.g. Pain Concern), international (e.g. IASP), and condition-specific groups (e.g. the Limbless Association).
You discuss with your patient the options for treatment for PLP, and he is keen to avoid multiple medications and side effects. You advise him to gradually titrate the gabapentin up to 900 mg three times a day, with advice to regularly consider its effectiveness and side effects. You reassure him that there are other medications that he can try if he feels gabapentin is not for him. You also suggest that, when he is on a stable dose, he tries gradually reducing his co-codamol, as this may be an unnecessary medication for pain relief, now his stump has healed.
You discuss other therapies, and he is not keen for treatments that involve touching his residual limb and so decides not to try the TENS machine you have suggested.
He has insight into his condition and appears motivated, so you refer him to a pain management programme for physiotherapy and possibly psychology. You give him advice about local organizations and some web-based resources, along with a leaflet about Airing Pain, the online radio station broadcasted by Pain Concern. He appears pleased and relieved that there are several helpful options and thanks you for your time.
Summary
PLP is pain perceived to be from the amputated body part. It is experienced in up to 80% of amputee patients and is often persistent. It is more common in patients with pre-existing pain and is thought to be due to several mechanisms, including nervous system changes such as central sensitization, remapping of the homunculus, and neuroma formation. Many strategies have been used for the prevention of PLP, although there is little evidence of effectiveness, except possibly for epidural anaesthesia. For longer-term management, there is more evidence for non-pharmacological treatments than for medication, and, in general, as for most pain patients, a biopsychosocial approach should be taken.
Alviar MJ, Hale T, and Dungca M (2011). Pharmacologic interventions for treating phantom limb pain. Cochrane Database of Systematic Reviews, 12, CD006380.
Flor H (2002). Phantom limb pain: characteristics, causes, and treatment. Lancet Neurology, 1, 182–9.
Jackson M and Simpson K (2004). Pain after amputation. Continuing Education in Anaesthesia, Critical Care & Pain, 4, 20–3.
Nikolajsen L and Jensen TS (2001). Phantom limb pain. British Journal of Anaesthesia, 87, 107–16.
Ypsilantis E and Tang TY (2010). Pre-emptive analgesia for chronic limb pain after amputation for peripheral vascular disease: a systematic review. Annals of Vascular Surgery, 24, 1139–46.