The response of patients with dystonia to pallidal procedures is not well understood. In this study, we assessed the postoperative outcome of patients with primary and secondary dystonia undergoing pallidotomy or pallidal deep brain stimulation.
Fifteen patients with dystonia had pallidal surgery (lesions or deep brain stimulation). These included nine patients with primary dystonia (generalized and cervical dystonias) and six with secondary dystonia (generalized, segmental, and hemidystonias). There were nine male patients and six female patients. The mean age at onset was 21 years for primary dystonia and 18 years for secondary dystonia. The primary outcome measure was a Global Outcome Scale score for dystonia at 6 months after surgery. Other outcome measures were the Burke-Fahn-Marsden Dystonia Rating Scale and Toronto Western Spasmodic Torticollis Rating Scale scores.
The mean Global Outcome Scale score at 6 months for patients with primary dystonia was 3 (improvement in both movement disorder and function). In contrast, patients with secondary dystonia had a mean score of 0.83 (mild or no improvement in movement disorder with no functional improvement). All patients with primary dystonia had normal brains by magnetic resonance imaging, whereas five of six patients with secondary dystonia had basal ganglia abnormalities on their magnetic resonance imaging scans.
This study indicates that primary dystonia responds much better than secondary dystonia to pallidal procedures. We could not distinguish a difference in efficacy between pallidotomy and pallidal deep brain stimulation. The presence of basal ganglia abnormalities on the preoperative magnetic resonance imaging scan is an indicator of a lesser response to pallidal interventions for dystonia.
Dystonia is a syndrome of sustained muscle contractions that produce twisting and repetitive movements and abnormal postures (10). The extent and severity of muscle involvement is variable, ranging from intermittent contractions limited to a single body region (focal) to generalized, involving the limbs and axial muscles. When the disorder is genetic or the cause is unknown and has dystonia as a sole feature, the disease is called primary or idiopathic. Alternatively, dystonia may be secondary to various brain insults, such as perinatal hypoxia/ischemia, encephalitis, traumatic diffuse axonal injury, and neuroleptic drugs. Dystonias can occur in isolation or, particularly in the case of the secondary dystonias, may be associated with other movement disorders, such as chorea, myoclonus, tremor, or other neurological deficits. Primary dystonia is often hereditary. In one specific form, DYT1 juvenile-onset primary dystonia (33), most patients have a tendency to develop generalized dystonia. This disorder is overrepresented in patients of Ashkenazi (central and eastern European) Jewish ancestry; the symptoms usually start late in childhood with dystonia in one limb and then may progress rapidly to generalized distribution (32), which, in its advanced state, is associated with muscle contractures and skeletal deformities leading to marked functional impairment.
Both primary and secondary dystonias, especially generalized and hemidystonia, have been notorious for their poor response to medical treatment. Patients with dystonia usually try many combinations of drugs, including anticholinergics; dopamine antagonists such as tetrabenazine, pimozide, and haloperidol; l-dopa; benzodiazepines; orally and intrathecally administered baclofen; and spinal cord stimulation. Botulinum toxin injections can provide symptomatic relief, but repeated use is not practical, particularly in children with generalized primary dystonia, because of the large doses needed for the many groups of muscles involved and the potential for eventual development of antibodies (11).
Stereotactic functional neurosurgery procedures have been tried for treatment of intractable dystonia for the past 50 years. Initially, thalamotomy was the procedure commonly performed. Several large series were reported between 1950 and 1980 (2,5,7,12,37,39). The results were difficult to interpret because various groups used different surgical techniques and nonstandardized outcome assessment measures, and the pathogenesis of the different dystonia syndromes was not always defined. The general impression of surgeons at that time was that the results of surgery were as frustrating as the medical treatment, particularly compared with the much more consistent results obtained in Parkinson's disease. Cooper (7) performed thalamotomies for more than 200 patients with various types of dystonia. He observed marked sustained improvement for 24.5% of patients and mild to moderate improvement for 45.2% of patients after a mean follow-up of 7.9 years. When Kelly (19) offered to review all of Guiot's cases of surgically treated dystonias, he was told not to waste time, because “mes résultats sont mauvais.”
Interest in the globus pallidus internus (GPi) as a possible target for the treatment of dystonia was rekindled with the rebirth of pallidotomy for treatment of Parkinson's disease in the early 1990s. This was triggered by the observation of the great efficacy of GPi pallidotomy in relieving l-dopa-induced dyskinesias, including dystonias (3,13,17,18,23,24,30). Several groups have reported their case series of pallidal lesions or deep brain stimulation (DBS) in patients with dystonia (14,15,25,26,29,43,44). The onset of improvement was often delayed, starting a few weeks after the operation in most of the patients, in contrast to the immediate relief of symptoms in primary dystonia.
In this study, we present the 6-month outcome in patients with primary or secondary dystonias who were operated on by a single surgeon (AML) in a 5-year period at the Toronto Western Hospital. The patients received pallidal DBS or pallidotomy. Some of these patients achieved remarkable improvement in symptoms associated with prominent gain in functional level. Others had disappointingly little benefit from surgery. The factors associated with better response versus lack of improvement are analyzed, and the emerging pattern is presented. Detailed reports of cervical dystonia and tardive dystonia patients are reported elsewhere (9a,9b).
PATIENTS AND METHODS
All patients with dystonia who received de novo pallidal lesions or DBS between October 1996 and March 2001 were included in this report. The patients were referred for surgery primarily because of ongoing disability in the context of failure of several regimens of medical treatment. The patients had preoperative magnetic resonance imaging (MRI) brain scans, neuropsychological testing, and neurological assessment and rating of the severity of the movement disorder and the degree of functional impairment. Patients who had previously undergone brain surgery were excluded.
The study group included 15 patients; 9 had primary dystonia, and 6 had secondary dystonia. Demographic characteristics and diagnoses of these patients are given in Tables 1 and 2. The mean age at onset was 21 years (standard deviation [SD], 18 yr) for patients with primary dystonia and 18 years (SD, 18 yr) for patients with secondary dystonia. The mean duration of dystonia before surgery was 10.4 years (SD, 7 yr) for primary and 11.8 years (SD, 9 yr) for secondary dystonia. The male-to-female ratio was 5:4 for primary dystonia and 4:2 for secondary dystonia. Of the nine patients with primary dystonia, six had generalized and three had cervical dystonia. Four patients with primary generalized dystonia had DYT1 dystonia. Of these, two (Patients 1 and 2) were siblings of Jewish ancestry (one previously reported ). Family history was also positive in one patient with cervical dystonia (Patient 8). Of the patients with secondary dystonia, three had hemidystonia and three had generalized dystonia, one of whom had Huntington's disease with combined dystonic and choreic features. All of the patients had tried more than four different lines of medical treatment, including anticholinergic drugs, neuroleptics, muscle relaxants, sedatives, and others, without benefit. The three patients with cervical dystonia had more than six sessions of treatment with botulinum toxin injections, each with initial benefit followed by secondary response failure.
All patients with a young age of onset had advanced degrees of dystonia with severe postural deformities that not only caused gross functional impairment with inability to carry out daily self-care activities but also impaired nursing care. The adult-onset dystonia patients were unable to maintain work because of their movement disorder and its accompanying pain. Patients with generalized or cervical dystonia had bilateral operations, whereas patients with hemidystonia underwent contralateral procedures.
The decision to offer pallidotomy versus pallidal DBS was based on several factors. Patients were preferentially offered pallidotomy if they required unilateral surgery, if they lived in foreign countries where postoperative follow-up and the ongoing need for stimulation parameter adjustments would pose logistical problems, or if they and their family stated that preference. Also, patients earlier in the study were more likely to have received pallidal lesions.
Generally, surgery was performed under local anesthesia in adults or propofol drip in children. A Leksell stereotactic frame was placed on the head, and stereotactic MRI was performed to identify the anterior and posterior commissures, as previously described (31). Microelectrode recording and stimulation were used to identify the sensorimotor portion of the internal segment of the GPi. Approximately 6-mm lesions were made in the posterior medial segment of the GPi. DBS electrodes were inserted within the sensorimotor GPi, 2 mm above the optic tract and 3 to 4 mm anterior to the electrophysiologically identified internal capsule.
Outcome Measures and Statistics
The primary outcome measure used was a Global Outcome Scale (GOS) score for dystonia at 6 months after surgery. In this scale, a score of 4 indicated marked improvement in movement disorder and function; 3, moderate improvement in movement disorder and function; 2, moderate improvement in movement disorder but minimal improvement in function; 1, mild improvement in movement disorder but no improvement in function; and 0, no effect. Other scales used in the assessment were the Burke-Fahn-Marsden Dystonia Rating Scale (BFMDRS) (4) for generalized dystonia and the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) (6,27,38) for cervical dystonia. Assessments were performed by the neurosurgeons and movement disorder neurologists. These were repeated in follow-up visits, and the GOS score (from 0 to 4) at 6 months after surgery was obtained.
Patients were grouped according to cause of dystonia: primary versus secondary; preoperative MRI findings: normal versus abnormal basal ganglia; procedure: lesion versus DBS; and other demographic criteria, including age, sex, and duration of dystonia. The outcome and mean GOS scores were then calculated for each pair of groups. The Mann-Whitney test was used to test the statistical significance of differences in outcome after surgery between groups of patients.
All nine patients with primary dystonia had normal preoperative MRI scans of the brain. Among patients with secondary dystonia, five had focal abnormalities involving the basal ganglia in the preoperative MRI scan (Fig. 1), and only one (with tardive dystonia) had a normal scan. Postoperative MRI scans confirmed placement of the lesions or electrodes in the posteroventral segment of the globus pallidus interna.
Fifteen procedures were performed in this group of 15 patients. Twelve patients underwent bilateral procedures. Three patients with hemidystonia had unilateral procedures. Seven patients had pallidotomy and eight had DBS as the primary procedure. One patient (Patient 11) had simultaneous insertion of GPi and ventralis oralis posterior thalamic DBS electrodes in an attempt to assess the relative merits of the thalamic and pallidal targets in treatment of his symptomatic hemidystonia.
Patients with primary dystonia showed striking improvements. The mean GOS score at 6 months for patients with primary dystonia was 3 (SD, 1.2) of a maximum possible outcome of 4. The mean percentage improvement in postoperative BFMDRS scores for generalized dystonia was 43% and that in TWSTRS scores for cervical dystonia, 58%. The GOS scores at 6 months after surgery and changes in other clinical rating scales are shown in Table 1.
Of the nine patients with primary dystonia, six showed marked improvement in their dystonia that started within weeks (1–6 wk) of the operation, and at 6 months, the outcome score was 4 of 4 in four patients and 3 of 4 in two. Of these patients, three had proven DYT1 genetic abnormalities and three had cervical dystonia.
The patient who fared the worst in the group of primary dystonia had a GOS score of 1 at 6 months. This patient (Patient 5), although there was no clear cause for his dystonia, demonstrated atypical features, including mental subnormality, speech abnormalities, marked bulbar dystonia, and a neonatal arachnoid cyst that needed shunting in early childhood. These features may suggest a subclinical perinatal hypoxic insult that later presented with dystonia at the age of 5 years. Two patients had moderate improvement in dystonic movements with little functional improvement. One is a 32-year-old DYT1 patient (Patient 4) who had only mild to modest improvement in gait and ability to feed herself, with an initial feeling of “relaxation” after bilateral GPi DBS insertion. However, during the subsequent weeks, the limb dystonic movements returned, particularly in the right side. MRI showed that one of the pallidal electrodes was positioned in the external rather than the internal pallidal segment. This may be the explanation for the lack of prominent improvement in her dystonia and functional state. The other patient (Patient 6) was diagnosed with adult-onset idiopathic dystonia at the age of 37 years. After insertion of bilateral GPi electrodes, she had temporary improvement but developed a malfunction of the hardware necessitating revision.
In contrast to the overall good response in patients with idiopathic dystonia, patients with secondary dystonia showed less and more variable benefit. The secondary dystonia patients had a mean GOS score of 0.83 (SD, 1.3) (mild improvement with no functional gain) at 6 months. The mean percentage of improvement in postoperative BFMDRS scores was 10.6%. The difference in outcome between the primary and secondary dystonia groups is significant (P = 0.01, z = −2.5). The percentage of primary and secondary dystonia patients in each GOS category is shown in Figure 2.
In the group with secondary dystonia, there was no improvement (GOS score, 0 at 6 mo) in the three patients with hemidystonia (Table 2). The child with glutaric aciduria Type I (Patient 13) presenting with severe axial and limb dystonias with postural deformities (preoperative BFMDRS score, 113) had only marginal improvement, with improvement in axial dystonia and relief of the arching posture allowing him to lie flat (BFMDRS score, 99 and GOS score, 1). The only patient who had consistent benefit (Patient 15) was a 51-year-old patient with bipolar schizoaffective disorder who developed severe tardive dystonia after many years of intake of antipsychotic drugs (long-term follow-up is reported in a separate case report [9a]). Notably, this patient had a normal brain MRI scan.
The appearance of the brain on the preoperative MRI scan was an important predictor of a good response to surgery. The mean GOS score for patients with normal versus abnormal MRI scans was 3 versus 0.4. The difference in GOS score between these groups was significant (P = 0.005, z = −2.8). The small sample size, however, does not allow us to distinguish the effect of having a normal MRI scan from having idiopathic dystonia.
We assessed whether lesioning versus DBS produced a greater benefit. The mean GOS score for patients with dystonia (primary and secondary) having pallidotomy was 2 (moderate outcome) and that for DBS, 2.75 (between moderate and good outcome). This was not statistically significant (P = 0.5, z = −0.6). When the age at surgery and duration of dystonia before operation were analyzed, we found that there was a tendency for better outcome scores in patients who were younger and who had an early onset of dystonia and shorter duration of disease before surgery. The mean GOS score at 6 months for patients with dystonia for 5 years or less before operation was 3, compared with a mean outcome of 2.1 for patients with more than a 5-year duration of disease. Here again, however, this finding did not reach statistical significance. There was no mortality or major neurological deficit after surgery in the series. However, 2 of the 15 patients (13%), both treated with bilateral pallidotomy, developed persistent speech impairment: hypophonia (n = 1) or dysphonia (n = 1).
Our results show that there can be a striking improvement in patients with primary dystonia undergoing pallidal procedures. The majority of our patients with primary dystonia had good to excellent outcome. In contrast, patients with secondary dystonia had minimal benefit, with a GOS score of 0.83.
The best results (GOS score of 4, marked improvement in movement disorder and function) were seen primarily in the patients with a DYT1 gene defect and in patients with focal cervical dystonia. One DYT1 patient (Patient 4) did not initially have the expected benefit, possibly because of suboptimal placement of her electrodes. This patient subsequently underwent reoperation, and the electrodes were inserted in a more medial position in the GPi. This resulted in marked improvement. Her BFMDRS score at 6 months after the revision was 21 (a 68% improvement from the initial preoperative score of 66). Thus, all DYT1 patients in our series have achieved GOS scores of 3 or 4. The particularly striking benefit in DYT1 after either pallidal lesions or stimulation was noted previously by our group and others (8,9,29,35,41). Surgery was not effective in patients with hemidystonia (GOS score of 0 of 4, no improvement, in the three patients). Other groups have also noted a tendency toward a better outcome in primary dystonia compared with secondary dystonia after pallidal lesioning (19) or DBS (9,22,28,36,40,42).
The use of pallidal DBS is also showing promise in the treatment of a specific form of primary focal dystonia, cervical dystonia. Several small series (1,16,20,21,34) have shown improvements of approximately 50 to 80% in the TWSTRS score and relief of associated pain in patients with cervical dystonia after pallidal DBS. This conforms to our results of between 43 and 82% improvement in TWSTRS score in three patients having bilateral GPi DBS.
The greatest benefits of pallidal surgery in this series were seen in the group of dystonic patients with normal brain MRI scans. Those patients had a mean GOS score of 3. In contrast, patients with abnormal MRI scans, all with secondary dystonias, had poor results, with a 6-month mean GOS score of only 0.4. From this observation, we conclude that brain integrity as assessed by structural MRI is a predictor of favorable outcome in dystonia patients undergoing pallidal procedures and that improvement after surgery is dependent on normal anatomic substrate.
Our data did not suggest a difference in the short-term outcome in patients receiving DBS versus pallidotomy. However, persistent speech disturbances occurred in two of the patients undergoing bilateral pallidotomies. This favors the consideration of DBS if bilateral procedures are considered, especially if performed simultaneously.
Statistical analysis performed in an attempt to isolate any other confounding factors by grouping the patients according to other demographic criteria, such as age at onset of dystonia, age at operation, duration of symptoms before operation, and sex, revealed nonsignificant differences across these groups. Statistical correction for multiple comparisons was not necessary, because most comparisons were nonsignificant.
A recurring observation in patients undergoing surgery for dystonia is that the benefit is often delayed and progressive. This is in striking contrast to patients with Parkinson's disease, in whom the benefits are seen immediately, indeed intraoperatively. The reason for this delay is not entirely clear. In primary dystonia, the dystonic symptoms are transient, less sustained, and often less severe. There may be musculoskeletal factors in patients with severe deformity that require time and training to improve. There may be a requirement for transsynaptic changes to occur to see maximal benefit. Finally, there may be a requirement for motor learning or relearning in patients who have not had normal motor function for prolonged periods of time. This motor learning may include reversal of maladaptive changes and functional or structural changes involving plasticity in neural circuits.
This study shows that primary dystonia, especially DYT1 and idiopathic cervical dystonia, responds better to pallidal intervention than dystonias secondary to destructive lesions and neurodegenerative conditions. Pallidotomy and pallidal stimulation can each be safe and effective treatments for dystonia.
We have found that a structurally normal brain, as seen on preoperative MRI scans, is a predictor of a good response to pallidal interventions. It is tempting to suggest that patients with primary dystonias have a functional disturbance in basal ganglia outflow that disrupts and enslaves its targets, the otherwise capable thalamocortical and brainstem motor systems. Removal of this disturbance, by either pallidal lesioning or DBS, allows the motor system to revert to a more normal level of function. As a corollary, patients with secondary dystonia and abnormal brain integrity do not have the neuroanatomic substrates to reconstitute normal motor function after removal of the abnormal basal ganglia outflow.
We thank Dr. Amr El-Naggar and Dr. Magdy El-Kalliny, Lake Cumberland Neurosurgical Clinic, Somerset, KY, for their continuing help and support.
This small series of patients seems to indicate that patients with primary dystonia have better results than those with secondary dystonia after posterior ventral pallidotomy or placement of deep brain stimulation (DBS) systems at that target. This is consistent with my own experience. With respect to secondary dystonia, I have sometimes been impressed with magnetic resonance imaging (MRI) scans on these patients showing a pathological lesion at the precise place where we would be making a therapeutic lesion and the patient, nonetheless, still had dystonia. This would indicate that the physiology of the motor abnormality may be different in primary and secondary dystonia. Alternatively, as the authors suggest, patients with secondary dystonia may not have the intact neuroanatomic circuits necessary for reconstitution of normal motor function after the reduction of pallidal outflow after pallidotomy.
Patrick J. Kelly
New York, New York
The role of pallidal surgery in dystonia is currently attracting great interest. The relative rarity and heterogeneity of this disorder preclude the publication of large, homogeneous clinical series. Although the population and surgical techniques described in this article were heterogeneous, the authors convincingly make the point that primary dystonia is a better indication for pallidal surgery than secondary dystonia at short-term follow-up times. I do not think much can be said about the relative role of pallidal lesioning versus pallidal DBS, given the lack of long-term follow-up for either procedure.
The authors make some tantalizing anecdotal remarks about the relation of electrode location to outcome. In one of their patients, a relatively lateral electrode (in the globus pallidus externus) was repositioned more medially, into the globus pallidus internus (GPi), with improved therapeutic effect. The exact location of DBS electrodes for optimal benefit in dystonia is not clear from the published literature and needs to be clarified by more detailed analyses of optimal electrode location.
From our own series, we can add to the authors’ cautionary stance with regard to secondary dystonia. We have observed that some secondary dystonia patients may become “stimulation-dependent” despite little overall functional improvement from DBS. In these patients, dystonic spasms may become worse than they were before surgery when the DBS is turned off or when the battery runs down. Such patients then require active maintenance of their DBS system in good working order just to maintain their preoperative status. We have not observed this “rebound” phenomenon in pallidal DBS for primary dystonias.
Philip A. Starr
San Francisco, California
Since the late 1990s, pallidal surgery (pallidotomy and DBS of the GPi) has attracted increasing attention as a treatment for dystonia. The majority of reports have been case reports and small series. More recently, case series of 5 to 25 patients have been published reporting outcomes on validated clinical rating scales with follow-up intervals from months to years. Most reported cases have had generalized dystonia, and a pattern is beginning to emerge relating surgical outcome to pathogenesis: primary dystonias, particularly genetic dystonias of the DYT-1 type, show greater and more consistent benefit than secondary dystonias with surgery.
In the past few years, reports of pallidal surgery for focal dystonia, particularly cervical dystonia, have begun to appear in significant numbers, and results seem to be favorable for this indication as well. Pallidal DBS is potentially advantageous to pallidotomy because it is reversible and more adjustable and because of the higher incidence of side effects with bilateral lesioning procedures. The increasing use of thalamic DBS instead of thalamotomy for tremor, as well as the use of pallidal and subthalamic nucleus stimulation for treatment of Parkinson's disease, is further driving the use of GPi DBS instead of pallidotomy for dystonia.
In this article, Eltahawy et al. report the Toronto group's results with pallidotomy and pallidal stimulation in patients with dystonia. Their results are consistent with previous reports. They used the heterogeneity of their patient population to investigate which types of patients do and do not benefit from surgery. All 15 patients who underwent pallidal surgery for dystonia at their institution between 1996 and 2001 were assessed 6 months after surgery with several instruments, including the Burke-Fahn-Marsden and Toronto Western Spasmodic Torticollis Rating scales. Their primary outcome measure was a 5-point global outcome scale.
Patients were grouped for analysis into “primary” dystonia (defined as isolated dystonia, without other abnormalities, either idiopathic or of genetic pathogenesis) and “secondary” dystonia (defined as dystonia with other associated features and/or caused by an identifiable nongenetic pathogenesis). The categories were heterogeneous. In particular, the primary dystonia group included six generalized (of which four were DYT-1) and three cervical dystonias, whereas the secondary group included three static lesions, two progressive neurodegenerative conditions, and one nondestructive lesion (tardive dystonia).
The length of follow-up was 6 months, and the results were good among the primary patients and poor among the secondary patients (52 versus 11%). Tables 1 and 2 of the article, however, reveal patterns of variation that are largely obscured by a division into primary and secondary pathogeneses.
Among primary dystonia patients, generalized DYT-1 dystonia had good results, similar to those of segmental cervical dystonia (64 versus 57% improvement), whereas non-DYT-1 generalized dystonia patients did considerably worse (20% improvement). Thus, the reported outcome for primary dystonia may be influenced by the number of non-DYT-1 generalized dystonia patients included in the study. Among secondary dystonia patients, the one patient with tardive dystonia did much better than the others in that group—in fact, better than some of the primary dystonia patients.
The authors point out that they cannot distinguish between pathogenesis (primary versus secondary) and imaging (normal versus abnormal basal ganglia on MRI) as predictors of outcome, because the two were almost perfectly correlated. A single patient (Patient 15, tardive dystonia) had secondary dystonia but a normal MRI scan, and that patient's outcome was more characteristic of the “primary/normal MRI” group, suggesting that the presence of an imageable lesion is the better predictor.
This study provides important information for centers already performing or considering DBS surgery for dystonia. However, readers should be aware of certain limitations. The patient numbers were small in each group, and there was no randomization between pallidotomy and DBS. The outcome assessment was relatively short (6 mo) and not blinded. It should also be emphasized that the programming of dystonia patients can be complex, with often delayed clinical benefits compared with Parkinson's disease and tremor.
In summary, this is an excellent study from an experienced group evaluating the usefulness of GPi DBS for the complex array of dystonias. DYT-1 dystonia and idiopathic cervical dystonia respond well to pallidal interventions, idiopathic generalized and tardive dystonia not as well, and dystonia caused by destructive lesions and neurodegenerative conditions poorly. Surgery seems to be more effective in structurally normal but functionally abnormal brains.
Ali R. Rezai
Dystonia is a terribly disabling disorder for which there is no truly effective medical therapy. Although pallidotomy had been used in the past for dystonia, most of the early surgery for dystonia focused on thalamotomy. Major studies of thalamotomy for dystonia suggested that primary dystonia could be helped, and to a lesser extent, secondary dystonia as well (1,9,10). For both, the degree of improvement was incomplete, with mild to moderate improvement that was gradually lost with time. Reoperations were typical of this time period. Multiple thalamic subnuclei were targeted, and none seemed to be more successful than others. Complication rates were relatively high. The “onion-skin” somatotopy of the motor thalamus, with distal to proximal segments of the contralateral limb represented in concentric layers, provides the anatomic basis for large lesions to encompass proximal body parts and thus a high risk of complication by encroachment on adjacent corticospinal tract and/or thalamic structures related to memory, speech, etc.
Pallidotomy offered an advantage over thalamotomy in that the motor segment can be lesioned with less likelihood of cognitive impairment. Despite this, very little surgery was performed in the years before the 1990s, when pallidotomy was revisited for Parkinson's disease. The observation that l-dopa-induced dystonic elements could be improved with pallidotomy led neurosurgeons to reinvestigate the use of pallidotomy for dystonia (6,8,10,11). Lesion location is a critical factor, and avoidance of associative territory in the globus pallidus is essential, just as it is in the thalamus. Bilateral symptoms and midline symptomatology in the neck, trunk, or other axial structures require bilateral procedures. Pallidotomy complication rates, even with bilateral approaches, seem to be lower than those for thalamotomy. A number of contemporary studies report improvement with pallidotomy for dystonia by standardized rating scores such as the Fahn-Marsden Dystonia Rating Scale and the Unified Dystonia Rating Scale (8,12).
The introduction of DBS offered an additional means of treating dystonia, as explored by several authors (2,4,5,7,12). Bilateral DBS has a very low complication rate and thus is preferred by many neurologists and neurosurgeons. A number of reports suggest that primary dystonia patients, and especially DYT-1 generalized dystonia patients, do extraordinarily well with lesioning or stimulation of the pallidum. Secondary forms of dystonia have shown improvement also, but far less than primary forms. Differentiation as to what is truly secondary and what is truly primary can be difficult. Lack of improvement in secondary dystonia is most likely attributable to the irregular pathogenesis and the potential for lesions outside the basal ganglia, which could be prohibitive to improvements.
The report by Eltahawy et al. that primary dystonia is more responsive than secondary dystonia, whether treated with pallidotomy or pallidal DBS, is therefore not particularly surprising. The secondary dystonias included one patient with Huntington's disease, for which there has been little surgical success (3). However, even excluding this patient, the results resemble those found by others. The weakness of this article is, of course, the short follow-up and the use of a global outcome scale rather than a standardized outcome instrument, such as the Fahn-Marsden Dystonia Rating Scale. Left unanswered are a number of important questions regarding the surgical treatment of dystonia. The patient population, as well as the surgical treatment, is extremely heterogeneous, leading to an inability to establish firm conclusions regarding lesions versus DBS, thalamus versus globus pallidus, or multiple other technical questions.
One of the major breakthroughs for advancing the surgical treatment of dystonia was the Humanitarian Use Approval on April 15, 2003, by the Food and Drug Administration. Medtronic Activa was approved for humanitarian device exemption for unilateral or bilateral GPi or subthalamic nucleus DBS placement for chronic, intractable primary dystonia, including generalized or segmental dystonia for patients 7 years old or older (http:www.fda.gov/edrh/ode/hdeinfo.html). This does require an Institutional Review Board-approved protocol. The result should be a more systematic study of dystonia allowing for many of the discrepancies to be resolved through formal scientific testing. National Institutes of Health funding for multicenter studies will be essential to resolve these issues.
Roy A.E. Bakay