|
|
| |||||||||||||||||||||||||||||||||||||||||
Background Posteroventral medial pallidotomy sometimes produces striking improvement in patients with advanced Parkinson's disease, but the studies to date have involved small numbers of patients and short-term follow-up.
Methods Forty patients with Parkinson's disease underwent serial, detailed assessments both after drug withdrawal ("off" period) and while taking their optimal medical regimens ("on" period). All patients were examined preoperatively, and 39 were examined at six months; 27 of the patients were also examined at one year, and 11 at two years.
Results The percent improvements at six months were as follows: off-period score for overall motor function, 28 percent (95 percent confidence interval, 19 to 38 percent), with most of the improvement in the contralateral limbs; off-period score for activities of daily living, 29 percent (95 percent confidence interval, 19 to 39 percent); on-period score for contralateral dyskinesias, 82 percent (95 percent confidence interval, 72 to 91 percent); and on-period score for ipsilateral dyskinesias, 44 percent (95 percent confidence interval, 29 to 59 percent). The improvements in dyskinesias and the total scores for off-period parkinsonism, contralateral bradykinesia, and rigidity were sustained in the 11 patients examined at two years. The improvement in ipsilateral dyskinesias was lost after one year, and the improvements in postural stability and gait lasted only three to six months. Approximately half the patients who had been dependent on assistance in activities of daily living in the off period before surgery became independent after surgery. The complications of surgery were generally well tolerated, and there were no significant changes in the use of medication.
Conclusions In late-stage Parkinson's disease, pallidotomy significantly reduces levodopa-induced dyskinesias and off-period disability. Much of the benefit is sustained at two years, although some improvements, such as those on the ipsilateral side and in axial symptoms, wane within the first year. The on-period symptoms that are resistant to dopaminergic therapy do not respond to pallidotomy.
Methods
Patients
Forty patients with idiopathic Parkinson's disease underwent posteroventral medial pallidotomy between June 1993 and January 1996 (Table 1). All the patients had initially had good responses to levodopa with the subsequent development of marked disability due to off-period immobility, disabling levodopa-induced dyskinesias, and increasing disability during the on period primarily from ambulatory disturbances, despite an optimal regimen of available antiparkinsonian drugs. Exclusion criteria included marked cognitive dysfunction, active psychiatric symptoms, concurrent neurologic or other, uncontrolled medical disorders, or previous brain surgery. Four additional patients underwent surgery but no lesions were made; these patients are not included in the analysis. Complications in these four patients included intracerebral hemorrhage requiring surgical evacuation (in one patient), unsuccessful microelectrode mappings (in two), and an exacerbation of paranoia (in one).
|
Patients were evaluated clinically as reported previously,2 with the use of a modified3 Core Assessment Program for Intracerebral Transplantation.4 Follow-up evaluations were performed at 1 week in the first 27 patients and at 3, 6, 12, and 24 months in 39, 39, 27, and 11 patients, respectively. When possible, postoperative doses of antiparkinsonian medications were maintained at the preoperative levels. Many patients, however, changed their medication doses because of reduced needs, particularly in the immediate postoperative period.
During the follow-up period, nine patients underwent additional surgical procedures. A repeated ipsilateral lesion was made in one patient after three months of follow-up. The initial lesion had been placed suboptimally, in a more anterior and dorsal position than usual, for fear of creating a visual deficit, because the optic tract could not be identified with certainty (the patient did not report seeing phosphenes in response to electrical stimulation at the location of the optic tract predicted by microrecording). In the second procedure, the optic tract was identified unequivocally with the use of light-flash–evoked action-potential microrecording, and the lesion was placed ventrally in a more appropriate location. After this operation, there was a marked improvement, especially in the patient's incapacitating dyskinesias. The other eight patients underwent contralateral procedures: pallidotomies in two patients (at 6 and 18 months), and implantation of deep brain stimulators in six (at 9 to 18 months). All eight patients had had improvement after unilateral pallidotomy and had requested a second procedure for persistent and disabling ipsilateral symptoms. These patients had further improvement after the second procedure, as reported elsewhere.5,6,7 For each of these nine patients, data on efficacy are included only to the last follow-up examination before the second surgical procedure, because including the subsequent follow-up data would have biased the results in favor of the initial procedure.
Surgical Procedure
The methods we used for microelectrode-guided posteroventral medial pallidotomy have been described in detail elsewhere.2,8 Microelectrode recordings and stimulation were used to identify the sensorimotor territory of the internal segment of the globus pallidus, the optic tract, and the internal capsule. One or two largely overlapping radiofrequency lesions, 6 mm in diameter, were produced with the use of a thermistor-coupled probe, 1 mm in diameter, with a 3-mm exposed tip. The temperature of the tissue at the lesion reached 60°C, 70°C, 80°C, and finally, 90°C for 60 seconds.
Statistical Analysis
Follow-up evaluations were performed at three and six months in 39 patients. The evaluation at six months was chosen for the assessment of early efficacy in the hope of reducing the influence of the placebo response, which should have been larger at three months than at six months. For the analysis of long-term efficacy, the patients were divided into two groups, those followed for one year (27 patients) and those followed for two years (11 patients), who were also included in the one-year group.
The primary measure of efficacy was the overall score on the Unified Parkinson's Disease Rating Scale1 (defined as the combined scores for activities of daily living and motor function, Parts II and III of the scale) in the off and on periods. The range of possible scores was 0 to 52 for activities of daily living and 0 to 108 for motor function, with a range of 0 to 160 for the combined score. Lower scores indicate better function. Secondary measures included the Schwab and England Activities of Daily Living Scale and subscores of the Unified Parkinson's Disease Rating Scale for tremor, rigidity, bradykinesia, postural instability and gait disorder, and dyskinesias, as defined previously.2 Pairwise comparisons between the results of the base-line evaluation and the results of the follow-up evaluations were made with Student's t-test or the Wilcoxon signed-rank test, as appropriate.
For the long-term groups, we used an analysis of variance with repeated measures or a nonparametric Friedman's analysis of variance with repeated measures on ranks. If no significant difference was found, a power analysis was performed. A correlational analysis was also used to detect any trend in loss of improvement over time (with a slope of zero indicating sustained improvement). In view of the number of repeated analyses, a P value of 0.005 was considered to indicate statistical significance, in order to avoid a type I error. To provide 95 percent confidence intervals, the mean of the percent change in scores for individual patients was used rather than the percent change in mean total scores.
A secondary hypothesis of the study was that pallidotomy would result in clinically significant improvement. This hypothesis was tested by examining the effects of pallidotomy on specific test items in the activities-of-daily-living scale that represent important daily self-care functions (feeding, dressing, and personal hygiene), two components of the motor section of the scale that are related to ambulatory ability (gait and postural stability), and the severity of contralateral dyskinesias as scored on the dyskinesia rating scale. For each of these items, a score of 2 or higher (out of 4) generally indicates some degree of dependence on others for care or a major interference in function. With the exception of the analysis of dyskinesias, this evaluation was applied only to the worst off period, since there were insufficient numbers of patients who were dependent on others in the on period to permit conclusions. Evaluations were performed at 6 and 12 months in the one-year group and at 12 and 24 months in the two-year group.
Results
Early Response
All 40 patients were followed for at least three months, and data were available at six months for all but 1 patient, who underwent a second procedure. Improvements in the off-period motor and activities-of-daily-living scores were in the range of 30 percent (Table 2). All off-period features of parkinsonism improved significantly on the side contralateral to the side on which surgery was performed. Ipsilateral tremor and rigidity did not change, but bradykinesia improved significantly. In the on period, there was a significant improvement only in the total activities-of-daily-living score (Schwab and England on-period scores showed a trend toward improvement [P = 0.007]). None of the motor subscores on the Unified Parkinson's Disease Rating Scale changed significantly in the on period. Tapping scores, a measure of bradykinesia, improved significantly on both the contralateral and ipsilateral sides in the off period and to a lesser but still significant extent in the on period. Mean (±SD) scores for contralateral dyskinesias were dramatically reduced (from 2.3±0.9 to 0.4±0.6, P<0.001), and scores for ipsilateral dyskinesias were reduced by 42 percent (95 percent confidence interval, 25 to 50 percent) (from 2.0±1.0 to 1.1±0.7, P<0.001).
|
Long-Term Response
Primary Measure
The off-period scores at each of the follow-up evaluations differed significantly from the base-line scores by an analysis of variance with repeated measures. There was no significant difference in the degree of improvement (the difference between scores at base line and each evaluation) over time (by an analysis of variance with repeated measures), suggesting sustained improvement, and the analysis demonstrated sufficient power. Regression analysis also demonstrated no significant change in the off-period scores over the follow-up period in either the one-year or the two-year group (mean regression slope, 0.065 [95 percent confidence interval, -0.047 to 0.117] at one year and 0.018 [95 percent confidence interval, -0.055 to 0.074] at two years). Three patients in the one-year group and one patient in the two-year group had slopes higher than the upper limit of the 95 percent confidence interval, suggesting some worsening of symptoms over time.
In the one-year group, the on-period scores at each postoperative evaluation were significantly different from the base-line score, with the exception of the score at six months (Figure 1). The analysis of variance with repeated measures showed no differences in improvement at different evaluations, and the analysis showed sufficient power. The mean regression slope was 0.009 (95 percent confidence interval, -0.075 to 0.151), indicating no change over time and a stable improvement in the on-period scores (approximately a 19 percent change). As shown in Figure 1, the improvement in the on-period scores was not sustained in the group of 11 patients followed for two years.
|
The significant improvements in off-period contralateral bradykinesia at one week (36 percent change; 95 percent confidence interval, 23 to 49 percent) and rigidity (51 percent change; 95 percent confidence interval, 32 to 68 percent) were sustained in both the one-year and the two-year groups. Improvement in tremor was sustained in the one-year group (53 percent change; 95 percent confidence interval, 35 to 71 percent). There was a tendency toward sustained improvement in the two-year group, but the changes in the scores were not statistically significant, because of the correction for multiple comparisons and the small number of patients (10) with tremor in this group. Significant improvement in the off-period composite score for postural instability and gait disorder was lost between 6 and 12 months in the one-year group and between 3 and 6 months in the two-year group.
Improvement in off-period ipsilateral bradykinesia was not sustained for more than three months in either group, and changes in the scores for other ipsilateral symptoms were not significant. As seen in the group of 39 patients at six months (Table 2), changes in the on-period scores were generally not significant or were unsustained.
Improvement in contralateral dyskinesias was sustained in both the one-year and the two-year groups. Trend analysis showed a slight worsening of contralateral dyskinesias between one and two years. The improvement in ipsilateral dyskinesias was sustained in both groups at one year, but the benefit was lost by the second year.
Level of Dependence
Between 44 and 52 percent of the patients who were dependent on help with activities of daily living (feeding, dressing, and hygiene) in the off period preoperatively were independent at the six-month follow-up (Table 3). This level of improvement was maintained at two years for feeding and dressing but was reduced at one and two years for hygiene. Improvements in gait and postural stability occurred in a smaller proportion of patients. In approximately 75 percent of the patients, on-period contralateral dyskinesias were reduced from a level of serious interference with function to a level that posed no interference. Additional data on outcome measures are available elsewhere (*).
|
There were no statistically significant changes in doses of individual antiparkinsonian medications or in the total dose of levodopa equivalents (Table 1) during the follow-up period.
Complications
Table 4 shows the complications of pallidotomy in the 40 patients. The one patient with mild preoperative dementia, who was enrolled early in the study period, had acute confusion and hallucinations in the immediate postoperative period. As they resolved, there was a persistent worsening in his level of cognitive dysfunction. This was the only patient in whom the disability from complications outweighed the improvement in motor function obtained from surgery. The persistent complications in the other 13 patients were generally rated as mild, and the patients and their families considered them worth tolerating for the benefit gained from the procedure. Included in this group was one patient who sustained a frontal venous infarct with persistent mild facial weakness, dysarthria, dysphagia with drooling, and behavioral changes. The transient bulbar, facial, and limb complications typically resolved within six weeks and were probably related to edema surrounding the lesion, which sometimes coursed along the internal capsule, as seen on postoperative magnetic resonance imaging scans. The persistent worsening of handwriting (in 4 of 15 patients with dominant-side lesions) was in striking contrast to the substantial improvements in all other manual activities. None of the patients had visual-field defects. Age and the side of the lesion were not correlated with the frequency or persistence of complications. All the patients underwent detailed neuropsychological testing, and many underwent formal behavioral evaluations. The preliminary results of these studies have been presented elsewhere.9
|
We present the results at six months in 39 patients with advanced Parkinson's disease treated with posteroventral medial pallidotomy, with long-term follow-up data in two overlapping groups: 27 patients followed for one year and 11 followed for two years. In this large prospective series, we used validated rating methods to assess signs and symptoms in both off periods and on periods, as well as on-period dyskinesias. Other studies have been retrospective,10,11,12 have used nonvalidated methods of assessment,10,11,12 have not obtained scores for defined off and on periods,10,11,12,13 have reported very inconsistent data,14,15 or have involved relatively small numbers of patients.16,17,18 Despite the advantages of our study, it was uncontrolled and unblinded, and the outcome measures derived from even the best clinical rating scales have a major subjective component.
Surgery resulted in a pronounced reduction in the score for levodopa-induced contralateral dyskinesias (82 percent; 95 percent confidence interval, 72 to 91 percent). This benefit persisted during two years of follow-up, although there was a mild worsening between the one-year and two-year evaluations. Ipsilateral dyskinesias improved but to a lesser degree, and this change was sustained for one year but was lost by the second year of follow-up. Both the activities-of-daily-living and the motor subscores in the off period improved by approximately 30 percent, and this improvement persisted for two years. The off-period subscores for tremor, rigidity, and bradykinesia improved significantly, and this benefit was sustained for two years. These changes were almost exclusively related to improvement in the contralateral limbs, although some short-term improvement in ipsilateral bradykinesia was also seen. On-period subscores showed little change, although the activities-of-daily-living scores did improve by 20 percent, possibly because of the striking reduction in dyskinesias. The off-period composite score for postural instability and gait disorder and subscores for gait, postural stability, and freezing were all significantly improved in the 39 patients followed for six months, but these improvements were not sustained. Smaller initial on-period improvement in these scores was also lost by three to six months. Complications were common but in most cases did not persist, were well tolerated, and were far outweighed by the benefit obtained from surgery. Two important exceptions were persistent dysphasia and hemiplegia due to an intracerebral hemorrhage in one patient without a lesion and permanent worsening of dementia in another patient.
Surgery resulted in an unequivocal clinical benefit. Approximately 50 percent of the patients who before surgery had been dependent on assistance in activities of daily living during the off period were independent six months after surgery, and this improvement was sustained for two years for feeding and dressing and partially declined at one year for hygiene but remained stable between one and two years. Improvements in gait and postural stability were less substantial but still evident in one third or more of the patients. Disability from dyskinesias improved markedly, and the benefit persisted for two years. These observations are particularly striking in view of the fact that the surgery was performed because no further clinical improvement could be obtained with adjustments in medication. As in the study by Baron et al.,18 age at the time of surgery affected the outcome with respect to off-period parkinsonism; the patients who were 60 years old or younger had a 36 percent improvement overall, whereas those 65 years old or older had only a 16 percent improvement. Although Laitinen et al. report that the dosage of levodopa can be reduced by 50 to 75 percent after surgery,10 most of our patients who were followed for one year or longer eventually required the same doses as before surgery.
The optimal candidate for pallidotomy remains to be identified. Obviously, patients severely disabled by asymmetric dyskinesias are excellent candidates for the procedure, since this feature appears to respond to surgery in almost all cases. However, this effect does not seem to be correlated with the degree of improvement obtained in other spheres. In general, the greatest improvement in parkinsonism is in the signs and symptoms that are most responsive to levodopa. Symptoms that persist in the on periods remain relatively unresponsive to surgery. Our group and others have obtained uniformly poor results in patients with other "parkinson-plus" syndromes that are not responsive to levodopa, such as striatonigral degeneration, progressive supranuclear palsy, and cortical basal ganglionic degeneration.19 Patients with marked preexisting cognitive dysfunction seem to have poorer responses than others18 and are at considerable risk for further cognitive decline as a result of the procedure (unpublished observations). We do not perform surgery in patients with impaired base-line neuropsychological function, because we believe there is a substantial risk of further decline to a state of disabling cognitive dysfunction.
It is unknown how long the effects of pallidotomy persist and what the long-term outcome is for patients who have undergone the procedure. A recent report by Fazzini et al. suggests that the benefit may be sustained for up to four years,20 although our observations of the loss of ipsilateral and axial improvement and the degree of overall improvement are in disagreement with their findings. It is unclear whether patients who have undergone pallidotomy will have a response to new treatments — for example, glutamate antagonists,21 fetal mesencephalic implantation,22 or regenerative therapies such as glial-derived neurotrophic factor.23 If high-frequency stimulation of the globus pallidus24 or subthalamic nucleus25 is shown to be at least as safe and effective as pallidotomy, it may make sense to offer this nondestructive approach, especially to younger patients, retaining the potential for a response to new treatments as they become available. However, practical issues, including cost and the large amount of time required to optimize deep brain stimulation (e.g., electrode combinations, frequency, amplitude, and pulse width), will strongly influence the final decision about surgery.
Until these questions are resolved, one can consider the use of pallidotomy in patients with late-stage Parkinson's disease who, despite optimal medical therapy, are disabled by levodopa-responsive off-period symptoms or levodopa-induced dyskinesias. Patients with predominant axial symptoms (freezing and falling), especially those that persist in the on period, with minimal dyskinesias should not be offered the procedure, nor should those with substantial cognitive decline or other levodopa-resistant parkinson-plus disorders.
Supported in part by grants from the Parkinson Foundation of Canada and the National Parkinson's Foundation. Dr. Lozano is a Medical Research Council of Canada Clinician Scientist.
We are indebted to Drs. J. Miyasaki and N. Galvez-Jimenez for valuable assistance in evaluating and caring for the patients; to Dr. J. Dostrovsky for electrophysiologic guidance; to Dr. A. Harsin for assistance in data analysis; and to Drs. K. Kieburtz, W. Koller, and N. Quinn for helpful comments on the manuscript.
Source Information
From the Divisions of Neurology (A.E.L., J.D.) and Neurosurgery (A.M.L., R.T., W.H.), Toronto Hospital, Toronto, and the Department of Neurology, University of Arizona, College of Medicine, Tucson (E.M.).
Address reprint requests to Dr. Lang at Toronto Hospital Movement Disorders Centre, 399 Bathurst St., MP 11-306, Toronto, ON M5T 2S8, Canada.
References
| |||||||||||||||||||||||||||||||||||||||||
Related Letters:
Pallidotomy in Advanced Parkinson's Disease
Ryan L., Weiner W. J., Lang A. E., Lozano A. M.
Extract |
Full Text
N Engl J Med 1998;
338:262-264, Jan 22, 1998.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
