A review of the medical literature on patients who seek amputation reveals 2 broad groups. In the first, patients with an apparently normally functioning limb without physical signs or pain seek amputation of the limb (apotemnophilia). In the other, patients experience chronic limb pain and seek amputation to remove the painful limb.
People Without Physical Signs Who Request Amputation—Apotemnophilia or Body Integrity Identity Disorder
The concept of body integrity identity disorder (BIID) is derived from case reports in the 1970s–'80s and 2 larger series of 52 and 20 patients10, 12 collected by Internet recruitment and assessed by telephone interview.
BIID affects both sexes, with a male predominance.10, 12 Affected individuals describe a sense of paradoxical incompleteness of their body because of the presence of an undesired limb, most frequently a leg, that does not match their “inner self body image.”10, 12 This compulsive feeling usually starts in childhood and worsens progressively with age, causing overwhelming anxiety and suffering in most cases.10, 12 Pretending to be an amputee (bending their legs, using crutches or bandages) usually provides transient relief. Over time, most subjects seek an amputation, inflict injuries, or try to cut the limb themselves. The affected body part is always healthy and not painful. A sexual paraphilic component (sexual arousal on fantasizing about being an amputee or by the vision of an amputee) is reported in a minority of cases.10 Medical and psychological interventions are thought to be ineffective, although partial relief of the symptoms may be achieved.13
Very little evidence is available on the underlying mechanism behind BIID. A neurological model14–16 based on dysfunction in right hemisphere areas that mediate body representation has been proposed. In the presence of normal sensory afferent input, a mismatch between central representation and peripheral feedback is hypothesized to lead to conflict in perception of the affected limb. The evidence for this model is limited, however. Skin conductance responses below the level of the desired amputation are lower than above,16 and it has been suggested that this reflects a congenitally dysfunctional right superior parietal lobule (with resultant abnormal body image) that causes abnormal sympathetic outflow via the insula.16 This is in keeping with the results of an interview study of 20 patients with BIID, where a left-sided predominance of symptoms and descriptions of the affected limb that were similar to those of patients with somatoparaphrenia led the authors to conclude that nondominant frontal/parietal structures might be involved in the pathophysiology of the condition.12
Our fixed dystonia cases have physical signs and symptoms and in this regard are clearly different from people with apotemnophilia. However, cases with apotemnophilia demonstrate how a deficit in central body schema might be implicated in a desire for amputation, a phenomenon rarely encountered in patients with neurological signs outside the setting of fixed dystonia and CRPS (see below).
People with Physical Signs and Pain Who Request Amputation
It is most uncommon for patients with disabling neurological diseases affecting the limbs in a focal or multifocal fashion, such as Parkinson's disease, stroke, multiple sclerosis, or neuropathy, to resort to amputation. In fact, there are only 3 case reports of this occurring in the medical literature, 1 due to a paralyzed limb after stroke,17 1 due to painful contractured legs after spinal cord injury,18 and 1 in a child with secondary dystonia,19 However, we acknowledge that other cases may exist and have simply not been reported (and were alerted by an anonymous reviewer of this article that he/she had experience of 2 cases of amputation for osteomyelitis associated with abnormal postures due to corticobasal degeneration and Rassmussen's-like encephalopathy).
At first glance, therefore, the 5 cases with fixed dystonia who sought amputation appear highly unusual in terms of previously reported cases of amputation in neurological disease. However, there are several reports of amputation (combined producing a total of more than 50 patients), mainly in the orthopedic literature, in patients with CRPS1.20–22 There is considerable overlap between fixed dystonia and CRPS1 and evidence that at least a proportion of patients reported in these case series had fixed dystonia. Nine patients in 1 series21 asked for amputation because of “ankylosed, hypersensitive fingers,” and 4 in another series22 had “reduced range of motion” of the affected limb that led to the operation. Most cases were longstanding and refractory to alternative measures, and a handful underwent multiple amputations. The main indications were intractable pain, functional disability caused by the painful limb, and recurrent infections of the limb. The largest series reported the results of 34 amputations in patients with CPRS1.21 Nearly 40% had periprocedural complications (infection and delayed healing), phantom limb pain occurred in 85%, CRPS1 recurred in the stump in a similar number, and 2 developed CPRS1 in another limb. Only 2 cases had benefit in terms of pain reduction, and 9 had some degree of improvement in function. The poor outcome in these CRPS1 cases following amputation is mirrored by the 3 fixed dystonia patients in our series who achieved amputation of their limb. It was universally unsuccessful, and phantom limb pain and spread of symptoms was the only outcome.
It is possible that BIID, CRPS1, and fixed dystonia share a common abnormality in central body schema representation. In patients with BIID, this abnormality may be developmental and arises without a specific triggering event. In patients with CRPS1 and fixed dystonia, a peripheral painful stimulus might act as a trigger to destabilize central body schema. This is supported in fixed dystonia by our recent data6 and in “pure” CRPS1, where mental rotation abnormalities that correlate with duration of symptoms and perceived intensity of pain are found.23 With regard to dystonia, mental rotation abnormalities have also been reported in DYT1 mutation carriers (both manifesting and nonmanifesting24) and primary hand dystonia,8 and therefore this abnormality is not specific to fixed dystonia. However, the coexistence of severe pain in patients with fixed dystonia may trigger or modulate body schema abnormalities in a different manner to primary dystonia, and in some patients this could lead to seeking amputation. One hypothesis is the interaction of a number of factors with a preexisting (developmental) abnormality in central body schema (as suggested in BIID). These factors include an underlying psychopathology that causes abnormal attentional focus on a painful injury, thereby increasing central plastic changes that occur normally after pain, voluntary or therapeutic immobilization of the limb (by patient or treating doctors, respectively), reducing normal afferent feedback from the limb leading to an abnormal increase in gain of excitability in central sensory areas. Many patients with stroke, neuropathy, or multiple sclerosis have painful and postured limbs, but do not seek amputation. In patients with fixed dystonia and those with CRPS1, the disorder appears to go beyond pain and posture to a deeper abnormal relationship with the limb, which in some leads to seeking amputation. This abnormal relationship might also extend to pursuit of other surgical interventions (cf, case 1), which are common in fixed dystonia. It would be possible to test these hypotheses, for example in functional imaging studies looking at response to painful stimuli with particular focus on nondominant parietal lobe structures and neurophysiological assessments of cortical excitability changes occurring with attention toward or away from the affected limb.
Amputations in CRPS1 and fixed dystonia are almost uniformly unsuccessful, as demonstrated by the medical literature and in our cases where amputation was achieved. Symptoms typically return in the stump or another body part, further supporting the notion of a disturbed central body schema. In view of this, we would strongly counsel physicians and surgeons against recommending or performing amputations in patients with CRPS1 and/or fixed dystonia, as the current weight of evidence is that such procedures are likely to do harm and are therefore unethical.
Fixed dystonia is a severely disabling condition affecting young adults. The theory outlined above generates some testable hypotheses regarding the pathophysiology of fixed dystonia that attempt to move beyond the dualistic battle between classifying these patients as organic or psychogenic toward a more integrated view of brain dysfunction in this enigmatic disorder.