NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Chronic Conditions (UK). Parkinson's Disease: National Clinical Guideline for Diagnosis and Management in Primary and Secondary Care. London: Royal College of Physicians (UK); 2006. (NICE Clinical Guidelines, No. 35.)

Cover of Parkinson's Disease

Parkinson's Disease: National Clinical Guideline for Diagnosis and Management in Primary and Secondary Care.

Show details

8Surgery for Parkinson’s disease

8.1. Introduction

Recognition of the limitations of dopaminergic therapy and the need to treat motor complications were the prime movers in the revival of functional stereotactic surgery for PD. This was aided by technological advances in the fields of imaging and computing. The introduction of CT and MRI scanning allowed surgeons to visualise and directly target deep brain structures without the need for indirect calculations from atlases based on cadaveric dissections. Modern engineering methods and computer technology resulted in easily used and reliable stereotactic hardware. Further advances came with the development of technology for deep brain stimulation (DBS), which has become the mainstay of movement disorder surgery.

Better understanding of the pathophysiology of movement disorders and of the basal ganglia circuitry has refined the surgical targets used in movement disorder surgery.

The ventrolateral nucleus of the thalamus has been one of the commonly used target sites for surgery in PD. Cells firing at tremor frequency can be identified in the ventralis intermedius (Vim) part of the thalamus and lesions or stimulators placed at this target can dramatically improve tremor.259

The serendipitous observation260 of the effects of accidental ligation of the anterior choroidal artery focused attention on the globus pallidus interna (GPi) as a target for surgery. One group261 identified the ventral and posterior parts of the internal segment (GPi) as the optimal site for surgical ablation. This group261 revived this procedure and it was in widespread use in the early 1990s. While pallidotomy significantly reduced dyskinesia, it had a lesser effect on tremor and akinesia. The morbidity of bilateral lesions and the introduction of subthalamic nucleus (STN) DBS reduced the use of pallidotomy. However, DBS of the pallidum has a role in dystonia and some patients with PD.

Figure 8.1. Structures of the basal ganglia.

Figure 8.1Structures of the basal ganglia262

(reproduced with permission from publisher).

Experimental studies using the MPTP primate model showed increased cellular activity in the STN, and lesions or stimulation of the STN can reverse the cardinal features of parkinsonism.263,264 However, surgeons were reluctant to lesion the STN in humans because of the risk of inducing hemiballismus. It was then shown that electrical stimulation of the STN-DBS265 produced dramatic improvement in parkinsonian symptoms in PD. STN-DBS has since become the most widely undertaken surgical procedure for PD.

Surgical techniques vary between centres, but it is generally performed in three stages: radiological localisation, physiological localisation, and then either an ablation or a stimulation procedure.

Radiological localisation involves the rigid fixation to the skull under local anaesthesia of a stereotactic base ring onto which a fiducial array can be mounted. In the past, ventriculography (ie outlining the ventricles of the brain by instilling air or contrast medium) was the radiological technique used, but this has been largely replaced by CT and MRI. It is now possible to identify most of the targets on MRI, and their position in stereotactic space is calculated using sophisticated computer programs.

When the radiological data have been acquired and analysed, the patient is moved to the operating theatre and the radiological localiser is replaced with a stereotactic arc system that allows the surgeon to pass electrodes through a small opening in the skull with a high degree of precision. This is usually undertaken under local anaesthesia to allow the surgeon to evaluate responses from the patient, though some centres now carry this out under general anaesthesia and depend on recording of cellular activity for final localisation of the target. Microelectrode recording of cellular activity is widely used for physiological localisation, but there is no consensus on the added value of this technique. Evaluating the patient’s response to electrical stimulation of the target usually makes further confirmation of accurate identification of the target.

When the target has been identified the options are of either using radiofrequency current for thermal ablation of the area or introducing a system for chronic electrical stimulation. Ablation has the advantage of being an inexpensive single procedure that does not require long-term follow-up for maintenance of implanted hardware. These advantages are largely negated by the irreversibility of the procedure and higher morbidity. Ablation has therefore largely been replaced by chronic DBS.

For DBS, the initial target localization is similar to that used for ablative procedures. Once the target has been identified the test electrode is replaced with an implantable quadripolar electrode, which is anchored to the skull. A period of stimulation using an external stimulator is sometimes used and when the efficacy has been confirmed the system is internalised. Under general anaesthesia fine cables are connected to the electrodes and tunnelled subcutaneously to a programmable pulse generator usually placed in the chest wall. The pulse generator is similar to a cardiac pacemaker with a high degree of programmability by an external device. It is possible to provide the patient with a degree of control of the stimulator. The pulse generator has a battery within it and depending on usage will have to be replaced in a simple surgical procedure every 3–5 years.

In view of the relative safety of stimulation procedures compared with lesioning, most surgery for people with PD today uses the former approach. The GDG felt therefore that it should confine its recommendations to STN, GPi and thalamic stimulation.

8.1.1. Methodological limitations of surgery trials

The included trials all had methodological limitations common to non-analytical study designs. Firstly, none of the included trials were randomised into surgical or non-surgical intervention groups. Secondly, none of the trials were performed under blinded conditions, either single or double. None of the trials were controlled with a cohort of non-surgical patients for longitudinal comparison over time.

There was also a general lack of inclusion/exclusion criteria, which could lead to pre-selected patient populations, lack of multi-centre comparative results analysis, and lack of sample size calculations. The mean follow-up of most trials was 7–12 months and the patient population tended to be younger with an average age of approximately 60 years.

What is the effectiveness and safety of any DBS procedure versus standard medical therapy in the treatment of motor complications in patients with PD?

8.2. Subthalamic nucleus stimulation

8.2.1. Methodology

No randomised or controlled trials were found on the effectiveness of any DBS procedure versus standard medical therapy. Therefore, the GDG agreed that large case series studies with a minimum sample size of 40 patients were to be accepted for review.

Nine papers were found which reported the effectiveness of STN-DBS versus standard medical therapy.

8.2.2. Health economic methodology

Four health economic studies met our quality criteria.266–269 One study267 evaluated the incremental cost-effectiveness of bilateral DBS of the STN or GPi versus best medical management. The study267 estimated the cost per QALY of bilateral DBS of the STN or GPi (intervention) versus best medical management in the US healthcare context.

Another study266 evaluated the incremental cost-effectiveness of STN-DBS versus drug treatment. This study266 estimated the extra cost per additional UPDRS point gained from bilateral high-frequency STN-DBS by comparing STN-DBS and drug treatment with drug treatment alone in the German healthcare context.

One study268 evaluated the costs of STN-DBS. The study268 estimated the total health service cost per patient including preoperative assessment, STN-DBS and postoperative management over a 5-year period in the UK healthcare context.

Another study269 evaluated the change in medication costs after bilateral STN-DBS. This study269 estimated the anti-parkinsonian medication costs pre- and post-operatively at 1 and 2 years after bilateral STN-DBS in a US healthcare context.

A simplified cost-effectiveness analysis of bilateral DBS-STN was estimated from the perspective of the NHS over 5-year period (Appendix F).

8.2.3. Evidence statements

With respect to quality of life:270

  • Parkinsonian symptoms, systemic symptoms, emotional functioning and social functioning all improved post-operatively (p<0.001).
  • The improvement in the score of UPDRS II correlated with the improvement in total Parkinson’s Disease Quality of Life (PDQL) score (p<0.001). (3)

With respect to efficacy, see Table 8.1.

Table 8.1. Bilateral STN stimulation (stimulator ‘on’).

Table 8.1

Bilateral STN stimulation (stimulator ‘on’).

With respect to predictive factors, the following results were observed (Table 8.2):

Table 8.2. Correlations between pre-operative and post-operative factors.

Table 8.2

Correlations between pre-operative and post-operative factors.

  • One study274 found: ‘the younger the age at the moment of operation and the shorter the duration of disease, the better the clinical outcome’. Another study271 reported: no significant correlation between age at time of surgery or disease duration and postoperative clinical outcome. (3)
  • One study275 found: UPDRS motor scores off medication were improved but less so in patients over 70 (<70 vs >70, p<0.02), and changes in UPDRS motor scores (on medication) worsened in patients over 70 and improved in patients under 70 (p<0.05). Another study274 found: no significant difference between patients older and younger than 60 years of age for UPDRS II, III and IV scores, and no significant difference in mean daily levodopa dosage at follow-up. (3)

With respect to adverse events, the following were reported following STN-DBS:

  • Neuropsychological events including: confusion, mania, delusion, depression, hypomania, aggressive behaviour, hallucinations, attentional and cognitive deficit, dementia, panic attack and apathy, which in some impaired activities of daily living.
  • Other adverse events including: hypophonia, transitory eye opening apraxia, thrombophlebitis, subcutaneous infection, haematomas, focal cerebral contusions, infections of the system (sic) (‘the system’ relates to the actual equipment used), dysarthria, disequilibrium, dystonia, weight gain, connection wound dehiscence, lead repositioning, air embolus, seizure and dyskinesias.
  • Stimulator-induced events including: electrode replacement due to unsatisfactory results, local pain at the implantation site of the pulse generator, reversible stimulation-induced dyskinesias after an increase in voltage, minor intracerebral bleeding at the site of the trajectory lead, dislocation of the impulse generator from site of implantation, transient paraesthesias associated with adjustment of stimulation parameter. (3)

With respect to withdrawal rates:

  • Two studies reported suicide attempts: one study reported patients with depression (three) who then attempted suicide (two)270 and the other study reported four patients who attempted suicide post-operatively (one died).273
  • In a third study,277 three patients died from causes unrelated to surgery or stimulation, and in a fourth study276 three deaths were reported (from intracerebral haemorrhage, myocardial infarction and suicide). (3)

8.2.4. Health economic evidence statements

Bilateral STN- or GPi-DBS costs an additional $49,194 in US$2000 (approximately £31,112) per QALY in comparison to best medical management.267 The study’s results suggest DBS may therefore be cost-effective if the quality of life after the procedure is improved by 18% or more compared with best medical management.

Bilateral STN-DBS costs approximately an additional DM1,800 (UK£580) in 2002 prices per unit improvement in UPDRS total score, derived from German costs and patient data.266 However, the costs will decrease further over the long term (> 1 year study period) from reduced drug expenditure and improved patient functioning. Therefore, the direct and indirect costs need to be assessed over the long term to sufficiently evaluate the cost-effectiveness of DBS.

The total health service costs of DBS of the STN, including pre-operative assessment, surgery and post-operative management over a 5-year period, was recently evaluated in the UK.268 The estimated total cost per patient was £32,526 for the bilateral procedure and £30,447 for the unilateral procedure (£ 2002).268

A US study evaluated the change in anti-parkinsonian medication costs 2 years after bilateral STN-DBS. The study found the medication costs had significantly decreased by 32% (p≤0.01) from the 1-year pre-operative costs and there was 39% reduction after 2 years.269 Pre-operatively, the average daily cost of PD medication was $19.53 ± 10.41 in US$ 2002 (approximately £11.92 ± 6.35) per patient. Post-operatively, this fell to $13.25 ± 5.41 (approximately £8.08 ± 3.30) per patient.269

The economic modelling performed for this guideline (Appendix F) suggests that STN-DBS costs approximately £19,500 per QALY over a 5-year period in comparison to standard PD care in the UK (£ 1998). The results are relatively robust based on one-way sensitivity analysis.

8.2.5. From evidence to recommendation

In the absence of RCTs, any conclusions on the efficacy and safety of bilateral STN stimulation must be tentative. Most of the patients in the open-label non-controlled trials described above were relatively young (aged around 60 years) so the results may not be generalisable to all those with the condition. Follow-up was for around 12 months only, which may not record later complications.

Despite these limitations, what evidence is available supports the efficacy of this technique in reducing off time, dyskinesia and levodopa dose, improving motor impairments and disability, and improving quality of life.

There is a small but significant risk of permanent neurological disability as a consequence of this operation, due mostly to cerebral infarction or haemorrhage. In a small number of patients, this can lead to death. Most other adverse effects of surgery were transient but concern remains regarding the incidence of neuropsychiatric complications, particularly depression and suicide. It is difficult to comment reliably on such issues in the absence of a control group.

The procedure requires an experienced, well-trained multidisciplinary team.

The high cost of this type of functional neurosurgery in PD is well recognised. No long-term data from clinical trials are available. However, economic modelling over a 5-year period performed as part of this guideline suggests that bilateral STN-DBS costs £19,500 per QALY in comparison to standard PD care in the UK (£ 1998).

The National Institute for Health and Clinical Excellence (NICE) published an Interventional Procedure Statement on bilateral STN stimulation in November 2003.23 This supported the use of the procedure provided normal arrangements for consent, audit and clinical governance are in place.

The PD SURG trial is evaluating the clinical and cost-effectiveness of STN surgery and recruitment is ongoing ( The NICE Interventional Procedure Statement encouraged clinicians to consider randomising patients in this trial.



Bilateral STN stimulation may be used in people with PD who:

  • have motor complications that are refractory to best medical treatment,
  • are biologically fit with no clinically significant active comorbidity,
  • are levodopa responsive and
  • have no clinically significant active mental health problems, for example depression or dementia. [D]

8.3. Globus pallidus interna stimulation

8.3.1. Methodology

No randomised or controlled trials were found on the effectiveness of any GPi-DBS procedure versus standard medical therapy. Therefore, large case series designs with a minimum sample size of 40 people were accepted for review.

8.3.2. Evidence statements

No trials were found which assessed the effectiveness of GPi stimulation in a case series with a minimum sample size of 40 people with PD.

8.3.3. From evidence to recommendation

While no RCTs or large case series have evaluated GPi-DBS, there are a small number of case series and comparative trials that suggest the procedure is effective (see section 8.4). However, it is likely to suffer from the same concerns regarding adverse events and costs as STN-DBS.

GPi-DBS is rarely performed for PD in the UK at present, though it is sometimes undertaken when STN-DBS is not possible.



Bilateral GPi stimulation may be used in people with PD who:

  • have motor complications that are refractory to best medical treatment,
  • are biologically fit with no clinically significant active comorbidity,
  • are levodopa responsive and
  • have no clinically significant active mental health problems, for example depression or dementia. [D (GPP)]

8.4. Comparison of different types of deep brain stimulation

What is the most effective form of DBS procedure in the treatment of motor fluctuations and complications in patients with PD?

8.4.1. Methodology

There were no randomised or controlled trials reporting the most effective form of DBS in the treatment of patients with PD. The majority of trials were retrospective case series, which compared the results of different techniques. Due to the lack of comparative trials in this area, the GDG agreed studies with a sample size minimum of 10 patients per arm should be reviewed.

Five trials278–282 were found which compared the before and after surgery results of STN-, GPi- and Vim thalamic DBS.

The majority of the patient population received bilateral implantation, though results were not reported separately from the unilateral implantation results.

8.4.2. Evidence statements

With respect to clinical efficacy

  • The following criteria were significantly (p<0.05) in favour of both STN- and GPi-DBS:
    • UPDRS I, II (off and on), III (off and on), IV278,279,281,282
    • – time in off state (UPDRS item 39)282
    • – Hoehn and Yahr scores281
    • – levodopa equivalent daily dose278,279,281
    • dyskinesia scores278,279
    • – patient and physician global assessments
    • – Schwab and England scale282
    • – home diary scores (% of time with good mobility and without dyskinesia during the waking day).279 (3)
  • The following criteria were improved in only one DBS technique versus another:
    • – Motor score improvement was more pronounced in STN patients than GPi patients (no p values stated).281
    • – Medication could be reduced only in STN patients and not in GPi patients (no p values stated).281
    • – Levodopa dose equivalent, though unchanged in the GPi group, was significantly reduced in the STN group (p=0.017).282
    • Trail making test (p=0.0013), test B (p=0.0015) and BDI (p<0.0001) improved under STN stimulation and not Gpi.281
    • – Literal (p=0.0018) and total (p=0.0002) fluency decreased under STN-DBS and not GPi-DBS.281
    • – Core Assessment Program for Intracerebral Transplantations dyskinesia rating scale favoured GPi (p=0.046) in absolute scores but percentage changes were not significant.282 (3)
  • Thalamic nucleus stimulation could not be compared directly to other techniques, as the outcome measures used to assess its efficacy are different from other techniques. The main outcome, tremor suppression, was found to be significantly improved with the procedure.283 (3)

With respect to adverse events, the following was reported:

  • No GPi-specific adverse events were reported.
  • See thalamic stimulation and STN stimulation sections for events specific to these procedures. (3)

8.4.3. From evidence to recommendation

There is no evidence from RCTs to compare STN with GPi stimulation. However, observational studies suggest that STN stimulation may lead to greater improvement in motor scores and more reduction in levodopa dose and depression scores. In comparison, GPi stimulation may lead to less cognitive impairment. Further work is required in this area.

It is recognised that pallidal stimulation for PD is rarely performed at present, though it is sometimes undertaken when STN-DBS is not possible.



With the current evidence it is not possible to decide if the STN or GPi is the preferred target for DBS for people with PD, or whether one form of surgery is more effective or safer than the other. In considering the type of surgery, account should be taken of:

  • clinical and lifestyle characteristics of the person with PD
  • patient preference after the patient has been informed of the potential benefits and drawbacks of the different surgical procedures. [D (GPP)]

8.5. Thalamic stimulation

How effective and safe is thalamic stimulation for the control of tremor in PD?

8.5.1. Methodology

Three papers284,283,285 reported the effectiveness of chronic stimulation to the Vim thalamic nuclei. The methodological limitations of these papers are similar to those of STN stimulation (see Section 8.2).

8.5.2. Evidence statements

With respect to tremor suppression:

  • All three studies284,283,285 showed a benefit of thalamic stimulation.
  • Only one study283 reported statistical analysis and stated that the following outcomes were significantly (p<0.05) improved: face tremor and observed tremor, hypokinesia, rigidity and ADL score. (3)

With respect to adverse events, the following were reported:

  • Post-operative events included: venous infarction with temporary aphasia, intraventricular haemorrhage and cardiovascular problems intra-operatively.
  • Stimulation-related events that occurred considerably more frequently in patients with bilateral implants (52%) as compared with unilateral (31%)285 included: dystonia, diplopia, sleepiness, altered mental status, paraesthesias, mild disturbance of gait and balance, mild dysarthria, increased drooling, nausea, insomnia, dysphagia, depression, wire tightness and dysarthria. (3)
  • No mortality was reported in any of the trials.

With respect to withdrawal rates:

8.5.3. From evidence to recommendation

There is no evidence from RCTs of the benefit of thalamic stimulation in PD. Data from observational studies suggest that this is an effective method of reducing tremor. The operation carries a risk of serious complications such as cerebral infarction and haemorrhage. The GDG recognised that this form of surgery is rarely performed for tremor in people with PD in England and Wales, having been superseded by STN stimulation.



Thalamic DBS may be considered as an option in people with PD who predominantly have severe disabling tremor and where STN stimulation cannot be performed. [D]

Copyright © 2006, Royal College of Physicians of London.

All rights reserved. No part of this publication may be reproduced in any form (including photocopying or storing it in any medium by electronic means and whether or not transiently or incidentally to some other use of this publication) without the written permission of the copyright owner. Applications for the copyright owner’s written permission to reproduce any part of this publication should be addressed to the publisher.

Bookshelf ID: NBK48525


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...