Imagined movements cause pain and swelling in a patient with complex regional pain syndrome.

Complex regional pain syndrome type 1 (CRPS1) is characterized by pain, swelling, and sudomotor and motor dysfunction. The affected limb is exquisitely sensitive, and gentle movements can exacerbate symptoms. Local or spinal mechanisms, or both, may mediate symptoms, including pain. Evidence of altered sensorimotor processing suggests that symptoms may also be mediated by cortical mechanisms.1 This clinical note reports on findings from one patient with CRPS1 in whom imagined movement of the affected limb, without local muscle activity or movement, increased pain and swelling. Case report. A woman aged 34 years sought treatment for CRPS1 of the left hand 17 months after an uncomplicated wrist fracture. Management involved performance of a motor imagery task in which she was shown 56 pictures of a hand in various postures (28 left hands and 28 right hands) and was required to imagine adopting the hand posture shown without moving her hand. After performing the task, the patient reported increased pain and swelling in the affected limb. The motor imagery task was repeated 7 days later. Before and after the task, which took ~3 minutes, pain intensity was assessed using a 10-cm visual analogue scale anchored with “no pain” and “worst possible pain,” and finger swelling was assessed using the circumference of the second, third, and fourth digits midway between the metacarpophalangeal and proximal interphalangeal joints (Beiersdorf–Jobst hand-measuring tape, Charlotte, NC). Measures were repeated 60 minutes after the task, during which time the patient sat quietly with her affected hand placed comfortably on her lap. The following day, to confirm that any effect of motor imagery was associated with the affected hand, the task was repeated, but pictures of the left (affected) hand were removed, and each picture of the right hand appeared twice in random order. Muscle activity during the task. To verify that there was no muscle activity during the motor imagery tasks, EMG activity was recorded using surface electrodes (DE-2.3, Delsys, Boston, MA; preamplified 1000 ; band filtered, 20 to 450 Hz) placed over the wrist extensors and wrist flexors and the biceps brachii of each arm. To verify that any effect was not caused by autonomic arousal, heart rate and galvanic skin response (GSR) were monitored. The patient was asked whether the task was stressful. During the task, there was no detectable EMG activity or increase in heart rate, although there was a GSR, which was consistent with low-level autonomic arousal, that commenced ~10 seconds before completion of the task. Pain and swelling were greater post-task but had returned to pretask values at 60 minutes. Thus, the data corroborated the patient’s report from 7 days earlier. This effect was not observed for performance of the second (unaffected hand only) task (figure). The patient indicated that neither task was stressful, which suggested the slight GSR observed in the first study was consequent to an increase in pain. Discussion. In this patient, imagined movements caused an increase in pain and swelling in the painful hand. The effect was not mediated by changes in muscle activity or by a stress response because there was no change in EMG or heart rate and only a late GSR, which was more consistent with a response to pain. Finally, imagined movements of the unaffected hand did not elicit the effect. Pain associated with imagined movements appears consistent with modern theories of pain that emphasize a common process underlying motor-behavioral outputs and pain, but those theories do not offer a neural correlate of that process.2,3 Possible mechanisms underlying the effect may involve the posterior parietal cortex, which contains representations of working body schema and is active during imagined and executed hand movements.4 This region also contains neurons that combine nociceptive and other sensory inputs to provide an integrated output pertaining to threat to bodily tissue.5 In a chronically active and sensitized nociceptive system, perhaps activation of the body schema for the affected part is sufficient to elicit such an output. That output could evoke swelling in the limb via projections to the amygdala, the internal circuits of which are thought to be capable of associating autonomic responses with specific behaviors.6 Alternatively, attention to the affected limb or preparing to move the affected limb may activate descending facilitatory projections from the rostroventral medulla to spinal nociceptive neurons.7 In summary, imagined movements caused an increase in pain and swelling, even though there was no muscle activity or movement of the limb. Although the mechanisms are not understood, this finding implies that symptoms of CRPS1 may be mediated in part by cortical mechanisms associated with movement of the affected part.