Tardive Dystonia

Updated: Sep 27, 2022
  • Author: Jonathan D Artz, MD, MS; Chief Editor: Glen L Xiong, MD  more...
  • Print

Practice Essentials

Tardive dystonia is one of several tardive syndromes, a group of movement disorders that occur relatively late in the course of ongoing treatment with dopamine receptor blocking agents. Tardive dystonia is characterized by involuntary muscle contractions and typically presents as abnormal posturing of voluntary muscles.

Signs and symptoms

Tardive dystonia is insidious in its development, often presenting first as a focal dystonia that becomes increasingly widespread over a period of months to years. Dystonia typically presents as repetitive or patterned movements that appear twisting or tremulous and are worsened with movement.

Dystonic symptoms can affect any region of the body with significant variability in presentation, with some typical presentations including the following:

  • Repeated jaw opening or closing or speech difficulties

  • Neck stiffness or pain

  • Uncontrollable blinking

  • Jerking or strange posturing of an arm or hand

See Clinical Presentation for more detail.


The diagnosis of tardive dystonia is a primarily clinical one based on observed phenomology and of course a history of neuroleptic exposure. Additional workup may be indicated if there are clinical features suggestive of other dystonia syndromes.

See Workup for more detail.


Even mild tardive dystonia frequently requires treatment. If possible, the first step in management should be to taper and discontinue offending neuroleptic agents to see if symptoms resolve. Unfortunately, discontinuation of neuroleptics can sometimes temporarily exacerbate symptoms (ie, withdrawal emergent symptoms). Discontinuation of treatment may be prohibited depending on the severity of underlying psychiatric illnesss. Although there is no data on long-term outcomes comparing patients who continue with treatment, clozapine and other antipsychotics with less potent dopamine blockade may be helpful.

Botulinum toxin is a useful intervention for focal dystonic symptoms. Due to it’s efficacy, rapidity of onset, and relative safety compared to most other interventions, it may be considered as a first-line treatment for severe tardive dystonia with focal symptoms.

Dopamine-depleting agents are currently the primary pharmacological treatment for tardive dystonia specfically. The VMAT2 inhibitors valbenazine, deutetrabenazine, and tetrabenazine are recommended. As with all causes of dystonia, anticholinergic medications, baclofen, and clonazepam are effective at improving symptoms. Medications with more limited evidence include amantadine, naloxone, levetiracetam, propranolol, vitamin E, and clonidine.

Deep brain stimulation (DBS) of the globus pallidus is the current treatment of choice for tardive dystonia refractory to medical treatment with alternative interventions such as DBS of the subthalamic nucleus currently having a smaller body of evidence.

Alternative treatments such as physical therapy have produced unclear results at this time.

See Treatment and Medication for more detail.



Dystonia is commonly defined as "a syndrome of sustained muscle contractions, frequently causing twisting and repetitive movements or abnormal postures." [6] Historically, the first use of the term was by Oppenheim in 1911, but earlier descriptions of the syndrome have been widely acknowledged. [7, 8]

The phenomenology of dystonia is remarkably variable. Differences in the extent and severity of muscle and frequency of symptom involvement range from intermittent contraction limited to a single body region to generalized dystonia involving the limbs and axial muscles. Features such as age of onset and presumed etiology play a tremendous role in prognosis and treatment. As such, a complete diagnosis of dystonia typically includes its characterization along three axes: age of onset, distribution, and presumed etiology. [7, 8]

Age is generally divided into early onset (≤ 26 years) and late onset (> 26 years), with a younger age of onset associated with a more generalized and severe course in primary dystonias.

Distribution is divided into focal (a single part of the body affected), segmental (contiguous parts of the body affected), and generalized (the entire body affected). Terms such as multifocal (multiple noncontiguous body parts affected) and hemidystonia (an entire side of the body affected) are also used.

An abbreviated list of body parts commonly affected can include all four limbs, the trunk (pisa syndrome for a lateral deviation, camptocormia for a severe anterior flexion), the neck (torticollis for lateral rotation, anterocollis for anterior flexion, and posterocollis for posterior flexion), the jaw (mandibular dystonia or oromandibular dystonia), the tongue (lingual dystonia), the vocal cords (spasmodic dystonia), the larynx (laryngeal dystonia) or the eyelids (blepharospasm). Symptoms can occur intermittently, only with specific tasks (such as writer's cramp, embouchur dystonia or golf yips), or more chronically. In general, the more of the body involved, the worse the prognosis.

The etiology of dystonias typically divide into 4 broad categories: primary, dystonia-plus, heterodegenerative diseases with dystonia, and secondary dystonia. Primary dystonia is used for familial and nonfamilial genetic syndromes where dystonia is the major feature. A dystonia-plus syndrome is also a genetic syndrome with dystonia as the primary symptom but with other neurologic symptoms prominent (such as the dystonia-Parkinsonism or dystonia-myoclonus syndromes). This is in contrast to heterodegenerative diseases with dystonia when dystonia is present but not the major symptom (such as Wilson's disease or PKAN). Secondary dystonia is a dystonia brought on by an inciting event, such as a stroke, trauma, or drugs.

Tardive dystonia is a form of drug-induced secondary dystonia. Persistent dystonia was introduced by the French to describe the late complications of chlorpromazine therapy. In 1973, Keegan and Rajput introduced the term dystonia tarda to describe drug-induced sustained muscle spasm causing repetitive movements or abnormal postures in patients who were treated with levodopa. [9]

Today, drug-induced dystonias are roughly divided into acute, chronic acute, and tardive. Acute dystonia is an immediate reaction to a drug treatment and chronic acute is the term used for continued symptoms with long-term treatment with an offending agent. In 1982, Burke et al coined the term tardive dystonia for dystonias that did not present as immediately after the introduction of the drug, but presented later and either continued or worsened after the drug's removal. [2] Tardive derives from the Latin word meaning late onset, and had already been used to describe abnormal orobuccal-lingual facial movements (ie, tardive dyskinesias) that also appeared as a late side effect to medications and tended to continue or worsen with the removal of the drugs.

In that paper, Burke and colleagues proposed the following four criteria for diagnosis:

  1. The presence of chronic dystonia

  2. A history of antipsychotic drug treatment preceding or concurrent with the onset of dystonia

  3. The exclusion of known causes of secondary dystonia by appropriate clinical and laboratory evaluation

The question of whether tardive dystonia should be considered a subset of tardive dyskinesia has been debated for a number of years. Grossly, there are many similarities. All tardive syndromes are caused by dopamine receptor blockers. They are all characterized by both their presentation days to months after the initial exposure and their continuation, or worsening, after the offending agent has been removed. However, in spite of these similarities, Burke et al suggested that tardive dystonia could be distinguished from the classic orobuccal-lingual choreic form of tardive dyskinesia not only by the dystonic nature of the involuntary movements but also by the frequency with which it causes significant neurologic disability. Burke et al noted that symptoms can begin after only a few weeks or a few days of exposure and the degree of improvement was much more limited compared with tardive dyskinesia. [2]

Other writers have followed the lead of Burke and his colleagues, publishing reviews that point to the differences in clinical manifestations, prevalence, prognosis, and treatments between tardive dystonia and dyskinesia. [10, 3]



The pathophysiology of tardive dystonia is not well understood. Due to this limited understanding, it is helpful to briefly review what is known about the pathophysiology of dystonias in general to put this information in context.

Dystonia is considered to be a sign of basal ganglia dysfunction. One line of evidence for this is from the stroke and traumatic brain injury literature. Dystonia never occurs with pure cortical lesions and only develops after striatal lesions, sometimes occurring weeks or months after the inciting basal ganglia lesion.

Electrophysiologically, dystonia is characterized by a sustained co-contraction of both agonist and antagonist muscles. Although most research has been done on primary focal dystonias, three areas of investigation have emerged in the literature. First, both EMG and imaging evidence shows a loss of reflex inhibition in spinal and brainstem reflexes and a loss of normal inhibitory patterns in the motor cortex. Second, there is evidence of abnormal cortical motor plasticity in patients with dystonia. Third, there is evidence of sensory processing abnormalities. Subtle impairment in spatial and temporal discrimination tasks as well as somatosensory evoked potentials are well documented. [7]

The pathophysiologic basis of tardive dystonia itself remains obscure. Why exposure to neuroleptics produces dystonia in some patients, chorea in some, and both in others is not clear.

Sachdev suggests that tardive dystonia may develop in individuals who are already vulnerable to dystonia, with the antipsychotic drugs activating a latent predisposition. [11]

However, although primary dystonias and tardive dystonias have many similarities, they also have differences and some have been hesitant to conclude that these exist on a continuum with each other. In terms of genetic studies, the evidence for similar genetic mechanisms has been lacking. For example, in many families affected by idiopathic torsion dystonia, a mutation of the DYT1 gene on band 9q34 has been identified, but currently, no evidence exists that similar genetic factors cause the predisposition to tardive dystonia.

Further, the genetic evidence has been lacking that factors that predict tardive dyskinesia also predict tardive dystonia. For instance, the Ser9Gly polymorphism in the D3 receptor has been associated with vulnerability to tardive dyskinesia, but a study by Mihara et al looking at that gene and two other mutations known to cause decreased metabolism of neuroleptics through changes in cytochrome P4502D6 and a decreased baseline density number of D2 receptors, respectively, found no overrepresentation with any of these mutations and their sample of nine patients with tardive dystonia. [12] To date, no genetic markers have been identified that predict the development of tardive dystonia.

The neuropharmacology changes underlying tardive dystonia also remain poorly understood. Dopamine receptor blocking agents can cause an acute dystonic reaction that appears superficially similar to tardive dystonia. Two basic theories have emerged to explain this reaction: hypoactivity of dopamine system leading to an overactivity of acetylcholine activity, and a paradoxical hyperactivity of dopamine due to preferential blocking presynaptic receptors. There are studies that support both of these hypotheses; however, it is unclear how well this can generalize to tardive dystonia. For instance, although clinically anticholinergics can be used to treat tardive dystonia, they are far less effective than they are in acute drug-induced dystonias.

One theory has been proposed by Trugman et al, who maintained that repetitive stimulation of the D1 receptor by endogenous dopamine, resulting in sensitization of the D1-mediated striatal output in the presence of D2 receptor blockade, is a fundamental mechanism mediating tardive dyskinesia and tardive dystonia. [13] This hypothesis is based on a relative segregation of outputs; the D1-mediated striatal output is directed preferentially to the globus pallidus, internal segment and substantia nigra, and pars reticulata, and the D2-mediated output is directed preferentially to the globus pallidus and external segment.

By selectively blocking D2 receptors, long-term treatment with a conventional neuroleptic disrupts the normal, coordinated balance of D1- and D2-mediated striatal outputs. With long-term neuroleptic administration, endogenous dopamine is able to stimulate D1 receptors, whereas D2 receptors are occupied by neuroleptics.

The hypothesis that sensitization of the D1-mediated striatal output is involved in the pathogenesis is consistent with both the delayed onset of dystonia after neuroleptic initiation and the persistence of symptoms after neuroleptic withdrawal; therefore, this model predicts that the D1 antagonist will be beneficial in the treatment of tardive dystonia.

The major limitation to this theory is that it tries to conceptualize tardive dystonia and dyskinesia with a single pathway, yet the 2 disorders have differences in epidemiology, natural course, and treatment.



Young age, male sex, mental retardation, and electroconvulsive therapy have been identified as specific risk factors. Exposure to dopamine receptor blocking agents (DRBAs) is essential for the diagnosis of this disorder. Antipsychotic medications are the most significant etiologic factor. Other medications associated with tardive dystonia include antiemetics (eg, prochlorperazine, promethazine, metoclopramide) and antidepressants (eg, amoxapine). Also, single case reports for veralipride, a benzamide derivative, and lithium causing dystonia have been reported.

Antipsychotic medications

The most common cause of tardive dystonia is exposure to antipsychotic medications (neuroleptics). Tardive dystonia develops in a shorter period and with significantly less total neuroleptic exposure than severe tardive dyskinesia. Also, patients with tardive dystonia seem to receive fewer doses of neuroleptic agents than persons who develop tardive dyskinesia.

All dopamine receptor antagonists that reportedly cause oral tardive dyskinesia also reportedly cause tardive dystonia. These include all first generation and second generation antipsychotic medications.  

The duration of exposure to antipsychotic medications required to cause tardive dystonia ranges from months to years. Exposure to antipsychotics need not be long, and a minimum safe period is not apparent. This minimum duration of antipsychotic exposure seems to be shorter for women. A longer duration of exposure to antipsychotic medication does not correlate with the severity of dystonia; however, patients with generalized dystonia have shorter neuroleptic exposure than patients with focal dystonia.

Other agents implicated in cases of tardive dystonia include amoxapine, an antidepressant with dopamine receptor–blocking properties, and antiemetics such as prochlorperazine, promethazine, and metoclopramide. [3]



The prevalence of tardive dystonia is 0.5-21.6% of patients who are treated with antipsychotic medications, with most on the lower end of that range. This condition undoubtedly is less common than oral-buccal-lingual tardive dyskinesia. In a survey of 555 psychiatric patients, Yassa et al found a prevalence rate of 34% for oral tardive dyskinesia and only 1.4% for tardive dystonia. [14] Similarly, Friedman et al found a prevalence rate of only 1.5% among 352 hospitalized psychiatric patients. [15] One study by Sethi et al indicated a prevalence rate of 21% for tardive dystonia among veterans institutionalized long-term. However, most of these cases were mild; only 20% were symptomatic. [16]

Tardive dystonia appears to occur in all ethnic and racial groups in which it has been studied. However, no large-scale prevalence studies have been done to determine its specific prevalence in each group.

The literature shows a higher prevalence in men than in women.

In 1982, Burke et al reported a 1.6:1 male-to-female preponderance ratio. In a follow-up of 107 patients, 16 of which had been previously followed by Burke, the ratio was 1.14:1. [2]

Friedman et al [15] and Yassa et al [14] conducted studies of two unselected psychiatric populations, the results of which supported a male-to-female predominance ratio of 4:1 and 3:1, respectively.

Although no large unselected population study exists, tardive dystonia appears to have an earlier mean age of onset than other related dystonic conditions.

In the study by Yassa et al, tardive dystonia had a mean age of onset of 40.5 years. [14] In a study by Kiriakakis et al of 107 patients with tardive dystonia, the mean age of onset was 38.3 +/- 13.7 years, with males having a younger age of onset then females (but also starting neuroleptics earlier). [4] It was also noted that the younger a patient's neuroleptic exposure, the shorter the interval before developing tardive dystonia.



The prognosis of patients with tardive dystonia is very poor. Unfortunately, once developed, this condition is usually persistent.

The discontinuation of all dopamine receptor antagonists appears to be the most important factor related to remission; patients who permanently discontinue these agents increase their chance of remission 4-fold compared with those patients who do not.

Another factor related to remission is the total duration of dopamine receptor antagonist therapy; patients taking dopamine receptor antagonists for less than 10 years have a 5-times higher chance of remission than those with more than 10 years of exposure.

Tardive dystonia is most likely permanent in patients who continue using neuroleptic drugs for more than 10 years. [4]

The indication for long-term use of dopamine receptor antagonists must be well established. Patients must be evaluated repeatedly in hopes of early detection of tardive dystonia; once tardive dystonia is present, the causative drug should be withdrawn if possible. If the patient is not disabled by dyskinesia, observing and hoping for a spontaneous recovery, rather than treating, is best.  


Tardive dystonia causes pain and physical and emotional disability. Disability is moderate to severe in 70% of patients with tardive dystonia.

Disabilities involve the activities of daily living and are socially embarrassing.

Impairment of speech, vision, eating, sitting, and gait has been reported. Pain is also often an accompanying symptom. Any truncal or lower-limb dystonia causes a gait abnormality, leading to a bedridden state only in severe cases.

The social embarrassment and distress over the movements are the issues that often concern the patients most. Limitations (real or perceived) in keeping gainful employment and making new friends and romantic partners can be devastating. [3]