In 1817, Parkinson, in the classic �Essay on the Shaking Palsy�, described a novel condition that now bears his name. He described the cardinal manifestations: tremor, bradykinesia, gait and postural disturbance. Charcot, who was so impressed by Parkinson�s description that he called the constellation of symptoms Parkinson�s disease, expanded description of the disease in the nineteenth century - named paralysis agitans by Marshall Hall in 1841. 

Parkinson�s disease accounts for 75% of cases of Parkinsonism .

The current surgical population is distinct from those treated before the use of L-dopa. Today�s surgical candidate is more likely to have true idiopathic PD and likely has been treated with several medications before surgery. For many, surgery is performed to eliminate l-dopa-induced dyskinesia, an iatrogenic condition that did not exist before the introduction of L-dopa therapy, and one that may be unavoidable no matter how L-dopa is administered. 

Physiologic basis of surgery:

The development of primate models of PD and the refinement of single cell microelectrode recording have resulted in a vastly improved understanding of PD pathophysiology. The following three points are essential to understand current surgical approaches:

1.      Although an intact pallidothalamocortical loop is necessary to exhibit tremor, lab data suggest that tremor activity is probably generated in the ventrolateral nucleus of the thalamus. These data support the clinical observation that lesions or high-frequency stimulation in this region best treat parkinsonian and essential tremor.  

2.      The striatum modulates activity in the globus pallidus pars interna(GPi) both directly and indirectly, via the globus pallidus pars externa(GPe) and the subthalamic nucleus(STN). In PD, loss of dopaminergic input to the striatum leads to a reduction in direct pathway activity and an increase in activity along the indirect pathway, changes that result in STN and GPi hyperactivity. The output from GPi  to ventrolateral thalamus is inhibitory, reducing the excitatory thalamic input to supplementary motor areas that are critical to the normal execution of movement. Thus hyperinhibitory outflow from GPi may account for the negative symptoms of PD(i.e., rigidity and bradykinesia). Reduction of this excessive inhibitory outflow, by lesioning or chronically stimulating GPi or STN, can reverse these symptoms.  

3.      With time, differential sensitivities of the direct and indirect pathways to L-dopa therapy may occur. This imbalanced response to exogenous L-dopa may result in choreiform movements that characterize L-dopa induced dyskinesias. It is not completely clear, however, why ablating GPi eliminates this condition.  

The advent of axial computed tomography and then magnetic resonance imaging has altered stereotactic technique. Employing these imaging modalities with modern stereotactic frames, the position of deep brain targets can now be sited directly, not estimated on ventriculograms. Nevertheless, current MR imaging does not always demonstrate deep brain structures (such as specific thalamic subnuclei) with sufficient resolution for it to be the sole means of targeting. Moreover, MR imaging is prone to distortions. Although typically small, they can be large enough to affect targeting for these types of procedures. Therefore some type of intraoperative physiologic confirmation is essential for performing successful movement disorder surgery.

Microelectrode, semimicroelectrode and macroelectrode techniques have been described. Microelectrode recording yields the highest quality localizing information, so that lesioning or DBS implantation can be performed with greatest confidence. Concerns that the increased number of trajectories employed with microelectrode recording increases the risk of hemorrhage have not been substantiated; however it is questionable whether the extra time, effort and expense of performing these are necessary to achieve results.

 Neuroablative procedures:

The goal of neuroablation is to disrupt irreversibly an abnormally functioning structure (usually a nucleus), leaving volitional movement intact. Ideally the lesion is just large enough to achieve the desired result but not so large as to cause collateral damage. During the first half of the twentieth century, neurosurgeons lesioned numerous sites throughout the motor system, searching for the ideal target. Lesions within the ventrolateral thalamus and GPi have yielded the best results. Many lesioning techniques have been employed � radiofrequency thermocoagulation is the most commonly used modality because of its reliability and simplicity.  

Ventrolateral thalamotomy: Hassler and Reichert are credited with making the initial foray into the ventrolateral thalamus, reporting improvement in both tremor and rigidity. Thalamotomy thereafter became the favored surgery in the pre-dopa era because the surgical response is instantaneous and more easily monitored than the response to pallidotomy. A number of surgeons have reported excellent long-term and short-term tremor suppression after thalamotomy, although contemporary authors do not report significant improvements in rigidity or bradykinesia. Tremor may be suppressed with lesions anywhere within the ventrolateral thalamus, but the ventralis intermedius subnucleus of the ventrolateral thalamus is considered by most to be the ideal target. The surgical mortality rate for thalamotomy is less than 0.5%; typically resulting from intracerebral hemorrhage. Morbidity may vary from 9-23% - but has decreased due to improved lesioning technique, and patient selection. Dysarthria and contralateral hemiparesis are the most common adverse events. Speech difficulties are even more prevalent in those undergoing bilateral procedures. Dystonia, hemiballism and athetosis have also been reported. In most instances, PD causes a resting tremor that diminishes with action, making this symptom one of the least disabling, and its elimination does not improve the functioning of the patient. Therefore it has given way to pallidotomy which addresses disabling PD symptoms such as rigidity, dystonia and dyskinesia.

Pallidotomy: in 1960 Svennilson et al reported superior results with pallidotomy by placing the lesion posterior, inferior and lateral to the previously described pallidotomy site. This was exemplified in the work done by Laitinen et al in 1992. Posteroventral pallidotomy dramatically improved rigidity, bradykinesia, tremor and ambulation in the majority of moderately advanced PD patients. Most important was the discovery that pallidotomy reduced or eliminated dyskinesias induced by dopa. It has been proved that pallidotomy reduces or eliminates dopa-induced dyskinesias, rigidity, muscular spasms and off state dystonia in patients with idiopathic PD who are still dopa responsive, but suffer from motor fluctuations. Tremor may also improve, but not as consistently as with thalamotomy. It improves the off state function and prolonged on states that are free of dyskinesia. Up to 60% improvements with effects lasting for 2 years have been reported.

The most frequent complication is visual field deficit (14%) � due to extension of the lesion into the optic tract. Injury to the internal capsule, facial paresis and cognitive deficits may occur. Complications increase with bilateral procedures � speech, swallowing and cognition.     

Gamma pallidotomy and thalamotomy: Leksell originally conceived the gamma knife to perform stereotactic neuroablative procedures without open surgery. But its use was limited by inadequate imaging. With high resolution MRI gamma knife surgery for neuroablation has become a reality. Only a few cases of thalamotomy and pallidotomy have been reported; the rate of improvement is lower than that for conventional procedures (50%). Disadvantage being that the patient has to wait for weeks to months to notice any positive changes. Therefore clinical efficacy/potential complications cannot be determined at time of surgery.

Deep brain stimulation: The idea for long-term DBS arose from the observation that high-frequency stimulation of the ventrolateral thalamus arrests tremor, a phenomenon that is employed for physiologic localization of the thalamotomy target. If one can arrest tremor with stimulation, why lesion? The advantages of DBS are � a reversible, functional lesion is made in lieu of a permanent anatomic one, stimulation parameters can be adjusted over time to maintain tremor suppression in the event of disease progression. The disadvantage being cost and maintenance of the device. The deep brain electrode is placed through a burr hole. As with the lesioning procedures the patient is awake and his neurologic status can be assessed throughout. Once placed correctly, the electrode is anchored to the skull. As the electrode is flexible, it can move with the brain and maintain its anatomic location. Each electrode has four contacts. Stimulation can be performed in monopolar bipolar fashion. Once acceptable stimulation parameters have been identified for the patient, they are programmed into the implantable pulse generator, which, similar to a cardiac pacemaker is placed within a subcutaneous pocket below the clavicle and connected to the electrode via wires that are tunneled beneath the skin. Stimulation parameters may be adjusted at any time employing a transcutaneous programmer. A magnet may be used to turn the stimulator off at night or to switch between two stimulator settings. The mechanism through which DBS achieves its functional results is presently unclear. Stimulation may create a depolarization block, jamming the signals emanating from an abnormally functioning structure, although this is unsupported by more recent findings. Alternatively antedromic and orthodromic propagation of the depolarization may affect distant structures that transmit to or receive impulses from the stimulation target.   Thalamic deep brain stimulation:  Initially Benabid et al employed thalamic stimulators contra lateral to previously successful thalamotomies in patients with bilateral tremor to reduce the risk associated with bilateral thalamotomy. The results demonstrated that thalamic DBS achieves tremor suppression with an equivalent efficacy to thalamotomy. The tremor suppression was dramatic and significant. Tasker retrospectively compared DBS and thalamotomy performed in one venter, and found that initial tremor suppression was comparable in both, but tremor recurrence was higher (15% versus 5%) in the thalamotomy group. Tremor recurrence necessitated reoperation in the thalamotomy group, while it could be resuppressed with adjustments to the stimulation parameters in the DBS group.  Neurologic complications were less frequent in DBS group and could be controlled with stimulator adjustments. Painful dysesthesias secondary to stimulation of the ventrocaudal nucleus (lies immediately posterior to the ventralis intermedius); dysarthria; and dystonia are the predominant complications. Hemorrhage rates and congnitive difficulties may be less frequent.

Radio frequency lesion generator and electrode.                         Ideal target location for Posteroventral Pallidotomy                    Ideal target location for Subthalamic nucleus.                       subthalamic nucleus as seen on T2W

Refinements in the understanding of the anatomic, physiologic and the neurochemical underpinnings of Parkinson's disease have greatly contributed to this renewed interest in neuroablative surgery and are giving rise to newer, more sophisticated therapies. Modern stereotactic technique permits targeting with millimeter accuracy. Deep brain stimulation has been introduced as an alternative to neuroablation. It is hoped that cellular transplantation or virally mediated gene transfer therapies that seek to restore or preserve function will soon be clinically viable.