Article Text

Download PDFPDF

Long term results of unilateral posteroventral pallidotomy for antipsychotic drug induced tardive dyskinesia
  1. M W P M Lenders1,
  2. H P J Buschman1,
  3. M D I Vergouwen2,
  4. E N H J Steur3,
  5. P Kölling4,
  6. M Hariz5
  1. 1Department of Neurosurgery, Medical Spectrum Twente, Enschede, The Netherlands
  2. 2Department of Neurology, Academic Medical Center, Amsterdam, the Netherlands
  3. 3Department of Neurology and Movement Disorder Unit, Medical Spectrum Twente, Enschede, The Netherlands
  4. 4Department of Psychiatry, Medical Spectrum Twente, Enschede, The Netherlands
  5. 5Unit of Functional Neurosurgery, Institute of Neurology, Queens Square, London, UK
  1. Correspondence to:
 Dr H P J Buschman
 Department of Neurosurgery, Medisch Spectrum Twente, PO Box 50 000, 7500 KA Enschede, The Netherlands;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Tardive dyskinesia (TD) is a well known side effect of neuroleptic drug treatment, and may coexist with tardive dystonia.1 It can be treated with several drugs, although they rarely lead to a complete removal of symptoms. The main treatment of TD consists of gradual neuroleptic drug dose reduction and where possible complete withdrawal.2 Nevertheless, in 40% of TD cases, symptoms do not disappear within 5 years of drug withdrawal. Wang et al first described a pallidotomy as a treatment modality for TD.3 In severe cases of TD and dystonia, this can lead to significant amelioration of abnormal movements and pronounced improvement in function and quality of life.4


We report a 51 year old, chronically hospitalised man with a 29 year history of schizophrenia. During this period, he was treated successfully for his psychotic symptoms with depot haloperidol and cisordinol. In 1993, he developed TD. Initially it consisted of an involuntary tic (myoclonus-like movements of the orofacial muscles), referred to in the literature as tardive tic.5 Both hands showed dyskinetic and myoclonic movements, the left hand less pronounced than the right, and there was severe acathisia. Haloperidol treatment was discontinued and symptoms of psychosis relapsed. Other medications tried included biperiden (6 mg/day), sulpiride (400 mg/day), amantadine (200 mg/day), dexetimide (1.5 mg/day), oxazepam (50 mg/day), and clonazepam (1.5 mg/day). Treatment with clozapine (up to 600 mg/day) and valproate (600 mg/day) improved the psychosis and the TD temporarily, but symptoms returned and became severely devastating and invalidating, because TD progressed to involve choreo-athetotic movements, most pronounced on the right side of the body and the midline. The patient also developed choreo-athethotic movements of the jaws, tongue, lips, neck, and upper chest muscles, leading to improper and irregular breathing, involuntary grumbling, production of uncontrollable noises, and frequent periods of pain in the chest. There was dysarthria, but no abnormal swallowing. He could walk, but only in a very dyskinetic manner. Socially he was totally isolated.

Huntington’s disease and Wilson’s disease were excluded. The pre-operative Unified Dystonia Rating Scale (UDRS) Revised score was 24. Dyskenesia and dystonia, as scored by the Unified Parkinsons Disease Rating Scale (UPDRS) were rated as 4 for both.

Owing to the unsuccessful medical treatment, and onset of a rapid deterioration in dyskinetic symptoms in the right side of the body, we decided on left sided posteroventral pallidotomy. In July 1999, the patient underwent surgery. Because of thoracic kyphosis, he could not be placed in a head coil or neck quad. We therefore obtained a pre-operative CT scan for “individual targeting”. The coordinates of a target in the posteroventral globus pallidus (GP) were established (2 mm in front of the mid anterior–posterior commissure (AC-PC), 5 mm below the AC-PC and 21 mm lateral of the AC-PC (head width 16 cm, third ventricle width 6 mm, AC-PC 25 mm long)). Five lesions were made at 8, 6, 4, and 2 mm, and on target at 82° for 60 s. As this did not result in a satisfying improvement of dyskinesia, a 2.0 mm diameter unipolar radiofrequency electrode (Fisher TCU 003) was inserted 2 mm lateral to the first target point. A second series of lesions was performed as above, which resulted in a complete disappearance of abnormal movements.

Post-operatively, no complications were observed. Involuntary grumbling was reduced by 95%. The patient was discharged from hospital 7 days after the operation, after withdrawal of phenytoin, dexamethasone, and ciprofloxacin. The post operative UDRS score was 2, and dyskenesia measured with the UPDRS was 0.

A post-operative CT scan of the brain showed a lesion centred 23 mm lateral of the midline in the calculated target point in the left GPi. One year after this pallidotomy, the patient was functioning very well. He was very content with the nearly complete abolition of TD, which allowed him again to paint, travel, and visit his family. Now, 5 years after the operation, he is still functioning very well, with no signs of recurrence of dyskinesia and dystonia. His UDRS and UPDRS scores remain at 2 and 0 respectively.


In this case report, we describe the outcome of posteroventral pallidotomy in a patient with drug induced TD. As more experience is gained,3,4 posteroventral pallidotomy, a procedure with documented effects on parkinsonian dyskinesias and dystonia, seems to be an effective treatment to improve or abolish symptoms of this movement disorder. The effect is not only apparent immediately and during the first months after pallidotomy, but as is shown in this patient can be maintained for long period of time.

Gpi target localisation was based on CT as an MRI was anatomically not possible. Ths would have been a good case for microelectrode recording, which at the time was not available at our hospital. Instead, we overlaid two CT images, taken parallel to the AC-PC line to enhance contrast, and compared them with the concurrent Schaltenbrand atlas page. This indicated that in the second, effective, series of lesions we targeted the Gpi close to the border of the Gpe. In Parkinson’s disease, very often a very small lesion in the Gpi can immediately be very successful; however in this our patient such a “parkinson GpI lesion” had no immediate visible effect. This does not exclude the possibility that over a longer period of time positive postoperative results could have evolved, as is often observed in patients with dystonia after pallidal stimulation or pallidotomy.

In our opinion, compared with pallidotomies in Parkinson’s disease, the target for TD and dystonia could be anatomically slightly different—that is, more lateral in the posteroventral GP perhaps because of different activity of the GpE. More study into this pathology is needed.



  • Competing interests: none declared

Linked Articles