Surgery for Parkinson's disease

 Introduction

There are between 5000 and 8000 British Columbians with Parkinson's disease (PD). The vast majority of them are well treated with the medications outlined elsewhere in this issue of the British Columbia Medical Journal. A small portion of them, however, have symptoms that are refractory to these medications. Some of these symptoms can be exquisitely treated with neurosurgical techniques.

There are a number of excellent reviews on the history of neurosurgery for PD.[1,2] In British Columbia, the surgical treatment of PD was first performed by Dr Peter Lehmann in the early 1960s. Before the advent of levodopa, there was little else to offer patients. Lesions were placed in the thalamus to help PD patients with tremor. The results, however, were variable because of the inconsistency of lesion size and location. After completing a fellowship in Europe studying stereotaxic surgery, Dr Ian Turnbull returned to Vancouver in 1966 and made many advances to the surgery for PD. He introduced the use of a stereotaxic frame that allowed more accurate target localization. He also switched the method of lesioning the brain from a leukotomy (which cut vessels as well as brain) to radiofrequency lesioning. Radiofrequency lesions are made by heating the uninsulated tip of an electrode to a defined temperature for a specified time. Lesion size can therefore be carefully controlled and adjacent blood vessels are not cut. Further reduction in patient morbidity followed the replacement of ventriculography with computer tomography for preoperative target localization in the 1980s. More recent advances have included MRI guided stereotaxis and a series of new operations utilizing deep brain stimulation.

The most common surgical procedures for PD in British Columbia and around the world are pallidotomy, thalamotomy, and deep brain stimulation. There are additional experimental procedures including transplantation of dopaminergic neurons (or dopamine secreting cells) and infusion of dopaminergic growth factors, all of which have great promise for the future but are not yet considered standard therapy.[3] This review focuses on the current surgical treatment for PD.

Pallidotomy

Before the advent of levodopa, there was little treatment for PD. Clinicians had recognized that PD patients who developed a basal ganglia stroke occasionally had improvements in their parkinsonian symptoms. Neurosurgeons therefore attempted to induce basal ganglia strokes in PD patients. The results were unpredictable and complications were frequent and severe. Initially the pallidum became the favored target and the operation, designed to make a small hole in it, was called the pallidotomy. Wycis and Spiegel reported that tremor was reduced in 78% of PD patients following pallidoansotomy.[4] Reports of reduced tremor and improved rigidity after pallidotomy followed from the United States, Japan, and Sweden. Fewer pallidotomies were performed after the introduction of levodopa and the realization that thalamotomy was better at tremor reduction than pallidotomy. The current era of surgery for PD was heralded by a report of excellent results for PD patients following pallidotomy in the posteroventral pallidum—a slightly different target than the previous standard.[5] This led to a resurgence of interest in pallidotomy. This new-found interest was bolstered by animal models that suggested Parkinson-like symptoms in primates were due to overactivity of the pallidum (inhibiting the motor thalamus) and that lesioning the overactive pallidum improved these symptoms.[6] A number of well-designed prospective studies on the effects of pallidotomy have now been published.[7-12]

 Indications and outcomes of pallidotomy

The ideal surgical candidate for pallidotomy would have idiopathic Parkinson's disease and symptoms that were refractory to medications. Since the most dramatic improvement after surgery is the reduction in contralateral dyskinesia, most surgeons select patients where this symptom predominates. Improvements are also seen in contralateral tremor, bradykinesia, rigidity, and overall balance and pain. Little long-term effect is seen ipsilaterally. Patients with motor fluctuations spend less time "off" and their "off" bradykinesia is improved following pallidotomy but their best "on" motor performance does not get much better (except for reduced dyskinesia). At 6 months post-operative, improvements in off-period UPDRS (Unified Parkinson Disease Rating Scale) total motor scores (17% to 46% decrease from baseline) and on-period drug-induced dyskinesia (43% to 92% decrease) have been reported.[7-12] Longer trials have demonstrated continued efficacy for off-period signs at 2- and 3-year follow-up.[13] The potential complications of pallidotomy are infection, seizure, and hemorrhage (common to all stereotaxic operations at approximately 1%) and injury to the internal capsule or optic tract. A number of articles on the neuropsychological effects of pallidotomy have shown subtle changes but no significant permanent detrimental effects.[14,15] Patients who are not good candidates for pallidotomy have Parkinson-plus syndromes or the major problem being speech, autonomic, psychiatric, or gait.

Thalamotomy

The results of placing a lesion in the thalamus, or thalamotomy, were first published by Hassler in 1954.[16] Throughout the 1960s different thalamic targets were lesioned until the ventral intermediate (VIM) nucleus emerged as the most effective target for tremor reduction. Since tremor can be quite resistant to medications, the thalamotomy procedure continued to be popular over the years. Unlike the pallidotomy procedure, however, the ventral intermediate nucleus thalamotomy has little or no effect on bradykinesia.

 Indications and outcomes of thalamotomy

Patients with unilateral, tremor-dominant PD can benefit from thalamotomy. Since most PD patients have other symptoms (bradykinesia) that respond better to pallidotomy, the thalamotomy procedure is not used as often. Reviewing the literature, Tasker found abolition of contralateral tremor in 45% to 92% of patients and less than 80% in his own series.[17] Complications specific to thalamotomy include contralateral weakness, arm and foot ataxia, dysarthria, and dysphagia. Patients who are not good candidates for thalamotomy have cognitive impairment, dysarthria, or PD symptoms other than tremor as the major problem.

Deep brain stimulation

The fastest growing area of PD surgery is deep brain stimulation. Instead of permanently lesioning an area, an electrode is placed in the target and used to "turn it off." There is no consensus on how deep brain stimulation works. It probably produces direct neuronal inhibition, but it may also cause stimulation of inhibitory pathways or "jamming" of neural networks. The electrode is connected subcutaneously, like a heart pacemaker, to a stimulator that sits below the clavicle. The benefit of this technology is that the area of the brain "turned off" is titratable. Office visits to change the stimulation parameters can optimize symptom relief or reduce unwanted side effects. The disadvantages to deep brain stimulation surgery include its cost (the BC provincial government budgeted for only 20 devices last year), the need for a general anesthetic to place the stimulator, expected battery failure in 3 to 5 years, and the added time required to fine tune the stimulation parameters to meet the patient's needs.

We are currently using deep brain stimulation surgery for patients who already have a lesion on the other side of the brain. Bilateral lesions can be dangerous, with the potential complications of dysarthria, dysphagia, and dementia. The deep brain stimulation allows bilateral surgery with the added safety factor that we can back away from a complication by turning down the stimulator. Deep brain stimulation is being used in three locations: the thalamic nucleus (stimulating the VIM only), the globus pallidus internus, and the subthalamic nucleus.

 Thalamic (VIM) stimulation

Tasker demonstrated a similar efficacy of thalamic stimulation and thalamotomy with fewer complications in the stimulation group.[18] Benabid and colleagues described VIM stimulation in 117 patients (80 PD, 20 essential tremor, and 17 other) with more than 85% left with little or no tremor in the contralateral arm.[19] Similarly to thalamotomy, VIM stimulation did not influence bradykinesia or rigidity. We are using thalamic stimulation in BC for PD patients with tremor who already have had a thalamotomy on the other side or who have tremor due to multiple sclerosis (where a lesion could spontaneously occur on the other side).

 Pallidum (GPi) stimulation

Siegfried and colleagues reported improvements in bradykinesia, gait, speech, and dyskinesia with bilateral globus pallidus internus stimulation in PD.[20] A prospective comparison of pallidal stimulation with pallidotomy in PD found the clinical benefits and safety comparable.[21] We use globus pallidus internus stimulation in BC for PD patients with dyskinesia and bradykinesia who have already had a pallidotomy on the other side.

 Subthalamic nucleus (STN) stimulation

The subthalamic nucleus sends excitatory (glutaminergic) signals to the pallidum and substantia nigra pars reticulata. These latter two nuclei are the inhibitory output of the basal ganglia. Reducing subthalamic nucleus activity with deep brain stimulation is postulated to improve bradykinesia by reducing the (inhibitory) output of the basal ganglia. Since glutamine has also been postulated to have a neurotoxic effect, reducing the subthalamic nucleus glutaminergic output is currently being studied to see if it can slow the progression of PD. Bilateral subthalamic nucleus stimulation in PD has been shown to reduce motor fluctuations (67%), dyskinesias (78%), and levodopa medications (61%).[22] We are using bilateral subthalamic nucleus deep brain stimulation in BC for patients with severe bradykinesia.

 Conclusion

In British Columbia, there has always been a close collaboration between the neurologists and neurosurgeons who treat patients with PD. Patients with symptoms refractory to medications have the potential to benefit from a variety of surgical procedures. In the future, there will be new treatments such as nerve cell transplantation, gene therapy, and growth factor infusion. Each new treatment will provide clinical benefits to our patients and an added understanding of Parkinson's disease.

 Acknowledgment

The authors would like to acknowledge Dr I.M. Turnbull for his dedication to this field of neurosurgery and to thank him for the information concerning the early history of this surgery in British Columbia.

 Further information

For further information about surgery for Parkinson's disease, visit the web site at www.interchange.ubc.ca/choney.

Christopher R. Honey, MD, DPhil, FRCS, and Ravikant S. Palur, MBBS, MS, MCh

Dr Honey is director of the Surgical Centre for Movement Disorders at the University of British Columbia. Dr Palur was the Functional and Stereotaxic Neurosurgery fellow at the University of British Columbia during the preparation of this manuscript.