We read with interest the article by Lim et al.1 about pain and dyskinesia in Parkinson's disease (PD). They proposed that similar mechanisms mediate both manifestations in PD. Here, we describe a case that provides further insight into the relationship between pain and levodopa-induced dyskinesia in PD.
A 48-year-old male physician developed right-hand rest tremor and bradykinesia as presenting features of PD, 11 years ago. “Wearing-off” motor fluctuations ensued some 5 years later under treatment with levodopa-carbidopa and subsequently, peak dose dyskinesia became problematic in the right arm.
Pain in the legs started 8 years after motor symptoms; this evolved toward increasing intensity over the next year and became generalized by gradual spreading from the legs to the trunk, over a few months, until reaching the upper limbs. Pain was maximal in the legs with a predominant “stocking-like” distribution at onset and was described as dull, burning, and not induced by tactile or stretching stimuli but showing high sensitivity to cold water. Pain was permanent but with fluctuations in intensity. The most painful periods coincided with “off” motor periods and mild improvement was noted in the “on” motor state. Evaluation of peripheral nerves and central somatosensory conduction was normal. All other diagnostic tests were normal. Treatment for pain was started with gabapentin up to 2,400 mg a day, which resulted in a mild improvement. Oxcarbamazepine, amitriptiline, tramadol, and carbamazepine were added sequentially, without additional benefit.
When we saw him, the patient reported severe pain during “off” periods. UPDRS-motor scale was 31 in the “off” (after 24 hours without medication) and 6 in the “on” and associated with moderate to severe dyskinesia predominantly in the axial and upper limb body segments. The pain was not associated with color skin changes or any other peripheral manifestations, but was clearly so invalidating as to interfere with his well-being and performance of daily life events. Pain abated considerably during “on” periods when severe dyskinesia was present; however, pain remained almost unaltered during “off” and “on without dyskinesia,” despite the fact that dystonic postures or rigidity were not present (Fig. 1).
After several attempts to control the condition with standard treatments, apomorphine subcutaneous infusion (9 mg/hour) was started. This caused a marked improvement in clinical condition and marked attenuation of “off” period and the severe pain. He continued to experience incapacitating pain at night and in the early morning and developed severe subcutaneous nodules related to apomorphine. Subthalamic nucleus deep brain stimulation (STN-DBS) was therefore indicated. After surgery and DBS adjustments, motor fluctuations and dyskinesia were markedly reduced, and the pain was almost absent most of the day, although L-dopa equivalent dose was not reduced significantly. A good degree of symptomatic control has persisted for more than 3 years.
This case exemplifies the presence of pain as a disabling nonmotor fluctuation, which was more incapacitating than the motor manifestations, as has previously been reported.2 In our patient, the threshold for pain response to dopaminergic stimulation was higher than the dopaminergic threshold for the “on” response and may be even above the dyskinesia threshold, as pain was only abolished when severe dyskinesia was present (“supra on” threshold). This different threshold could make the recognition and proper management of this nonmotor fluctuation difficult.3 In our patient, the combination of dopaminergic drugs (which remained unchanged) and DBS may have had a synergistic interaction, as has been shown for other parkinsonian features.4
The case reported here exemplified the possibility of pain in PD not directly related to abnormal muscular activity, such as “off” period dystonia. We would therefore like to suggest that in our case, pain has a central origin and has the clinical features of neuropathic pain. The mechanism by which pain may arise in PD despite adequate treatment for control of motor features is obscure.
Dopamine has been postulated as a modulator for the processing of painful experience, particularly in the mesolimbic system originating in the ventral tegmental area.5 A lower pain threshold and more extensive cortical activation induced by pain (measured by H215O PET) have been recognized in PD patients, suggesting a link between dopamine loss and hypersensitivity to pain stimuli.6 This is in keeping with a role for the basal ganglia in sensory processing. In fact, there is experimental data showing cortical somatosensory (including nociceptive) input to the striatum and physiological activation of striatal regions in response to nociceptive peripheral stimulation. In the DA depleted state, the striatum may fail to adequately select out sensory afferent activity, giving rise to pain.7 Why this is only present in some patients is unclear, but this should be included as part of the clinical spectrum of PD.