Bilateral pallidal stimulation for cervical dystonia: Dissociated pain and motor improvement

Patient 1.

A 35-year-old woman presented with a 20-year medically intractable segmental ICD and severe pain in her neck and shoulders. At age 34 she experienced reactive depression and attempted suicide. Results of blood analysis—including copper, ceruloplasmin, and acanthocytes—cranial MRI, and muscle biopsy were normal. She received various unsuccessful trials with diazepam, trihexyphenidyl, botulinum toxin, and opioid and nonopioid analgesics. Depressive symptoms improved only mildly with fluoxetine. Preoperative examination disclosed dysphonia, left torticollis and laterocollis, left shoulder elevation, and dystonic postural tremor in both arms (torticollis rating scale of Tsui et al.4 [RST] total score = 21; Hamilton depression score = 36). Palpation over dystonic muscles reproduced the pain that brought the patient to the clinic. Mean pain intensity was rated daily in the week before surgery with a 100-mm visual analogue scale (VAS).5 The reading was 84 mm (range 70 to 90), with pain present more than 75% of the waking day. One week after initiation of GPi DBS, the patient showed a remarkable improvement of pain and dystonic tremor of the upper limbs. Weekly rated VAS scores maintained lower values during the 24-month follow-up (20 mm; 64% improvement; table). A gradual improvement (21%) in cervical dystonia was observed within about 3 months ( see the table). A persistent improvement of mood was also noted (greater Hamilton score at follow-up = 12). Accidental disconnection of the stimulators and a formal attempt to switch the stimulators “off” were followed by pain aggravation; alleviation was observed a few hours after switching the stimulators “on.”

Patient 2.

A 65-year-old woman with hypertension presented with a 7-year medically intractable ICD and severe pain in her neck and right shoulder. Results of complete blood analysis were normal. Brain MRI showed minimal leukoaraiosis and cervical MRI showed moderate arthrosis at C5-C6 level. She received various unsuccessful trials with benzodiazepines, trihexyfenidyl, botulinum toxin, and opioid and nonopioid analgesics. Preoperative examination revealed antecollis, right laterocollis, and left torticollis (RTS score = 17). Palpation over dystonic muscles reproduced the pain that brought the patient to the clinic. Mean VAS pain intensity was rated daily in the week before surgery. The reading was 77 mm (range 60 to 90) with pain present more than 75% of the waking day. A remarkable alleviation of pain (present less than 25% of the day) was observed one week after GPi DBS initiation. Weekly rated VAS scores lowered to 20 to 30 mm during the entire follow-up (55% improvement; see the table). Cervical dystonia improved gradually within the first 3 months (7% improvement, RST = 15, see the table) but increased to presurgical level (RTS score = 17) at 6-month follow-up. Botulinum toxin injections were needed to maintain postsurgical gains. Three attempts to switch the stimulators “off” for 72 hours were followed by pain aggravation after approximately 24 hours, and pain alleviation occurred a few hours after switching the stimulators “on.”

Discussion.

Although peripheral–central interactions should be considered, the frequent alleviation of dystonic pain after botulinum toxin suggests a mainly musculogenic origin of pain, probably related to the abnormal involuntary activities of the neck muscles.6 Thus, substantial reduction in the frequency and severity of dystonic contractions after bilateral pallidal DBS is a likely explanation for concurrent pain alleviation in some patients.2 However, the modest motor gain of our patients (at the very low range of improvement following pallidal surgery, varying between 34 and 79%)1,2⇓ contrasted strikingly with the marked decrement in their appreciation of pathologic pain at final follow-up: 64 mm of improvement in VAS with 21% motor improvement in Patient 1; 55 mm of improvement in VAS associated with a transitory 7% motor improvement in Patient 2. This suggests that, at least in some dystonic patients, dysfunction of basal ganglia thalamocortical circuits can result in deficient processing of nociception, partially contributing to pathologic pain generation or maintenance and to the observed changes in pain perception in ICD (as well as in other basal ganglia disorders such as PD).7

The wide dissociation between pain and motor improvement in our patients suggests that dysfunction of motor and somatosensory circuits in dystonia do not always proceed in parallel. Indeed, DBS may have induced different changes in the frequency or on the pattern of neuronal discharge of segregated populations of GPi neurons, producing dissociated pain and motor responses.

Considering the detrimental influence of pain in the expression of dystonia, resetting of a putative dystonic pain circuit by DBS might largely or exclusively account for the limited motor improvement in the current and other patients (Patient 1 of reference 2). Worsening of dystonia with re-emergence of pain after withdrawal of DBS in our patients supports this view. Finally, the marked and long-lasting improvement in pain and its re-emergence when DBS was turned “off” outweigh a placebo effect as the main cause of pain relief.

Pallidal DBS may be valuable for selected patients with severe dystonia and pain.