Tardive Dyskinesia
Tardive dyskinesia (TD) is a hyperkinetic movement disorder characterized by repetitive, stereotyped, purposeless involuntary movements — most classically of the orobuccolingual region — that develops as a delayed consequence of exposure to dopamine receptor-blocking agents (DRBAs). First recognized in the 1950s following the introduction of chlorpromazine, TD remains the most common and best-studied tardive syndrome, affecting approximately 20–30% of patients on long-term antipsychotic therapy. The 2017 FDA approval of VMAT2 inhibitors (valbenazine and deutetrabenazine) has fundamentally changed the treatment landscape.
Bottom Line
- Prevalence: Global mean ~25% across all DRBA-treated populations; 30% with first-generation antipsychotics (FGAs) vs. ~21% with second-generation antipsychotics (SGAs); massively underdiagnosed in clinical practice (95% of cases in one EHR analysis had no ICD-coded diagnosis)
- Pathophysiology: Multifactorial — dopamine receptor supersensitivity, oxidative stress/excitotoxicity, GABAergic neuronal degeneration, and maladaptive striatal neuroplasticity
- Prevention is paramount: Use lowest effective dose for shortest duration, prefer SGAs (especially clozapine), monitor with AIMS at baseline and every 3–12 months depending on risk
- VMAT2 inhibitors are first-line treatment: Valbenazine 40–80 mg/day (KINECT trials) and deutetrabenazine 12–48 mg/day (ARM-TD, AIM-TD) both have Class I evidence
- TD is frequently irreversible: Spontaneous remission occurs in only ~2.5% per year while on the offending agent; even after drug withdrawal, many patients have persistent symptoms for years
Epidemiology
Prevalence
| Population | TD Prevalence |
|---|---|
| Global mean (meta-analysis of 41 studies, 2000–2015) | 25.3% |
| Current FGA treatment | ~30% |
| Current SGA treatment | ~21% |
| FGA-naive patients on SGAs only | 7.2% |
| SGA-treated with prior FGA exposure | 23.4% |
| Intellectual disability on antipsychotics | 42–45% |
Incidence Rates
| Population | Annual Incidence |
|---|---|
| FGA-treated patients | ~6.5% per year |
| SGA-treated patients | ~2.6% per year (range 0.8–5.3%) |
| Patients >50 years of age | Up to 30% after 1 year of exposure |
The risk ratio for SGAs vs. FGAs is approximately 0.47, with a number needed to treat of 20 (i.e., 20 patients need to be treated with an SGA instead of an FGA to prevent one case of TD).
Natural History
- Spontaneous remission while continuing the offending agent: ~2.5% per year
- After DRBA discontinuation: up to 33% achieve remission within 2 years; most remissions occur within the first year
- One long-term study (42 patients, 6.7 years): only 12% achieved overall remission; drug discontinuation increased remission 4-fold (22% vs. 5.8%)
- Age strongly influences remission: TD in patients <60 years improves 3x more often than in those >60
- A university movement disorder clinic study found tardive syndromes are rarely reversible after discontinuing DRBAs, underscoring the chronic nature of the condition
Massive Underdiagnosis
A 2025 EHR analysis of >32,000 antipsychotic-treated patients found that while 1,301 (4.0%) had evidence of TD, only 64 (4.9% of those) had a documented ICD-coded TD diagnosis — a 95% underdiagnosis rate in clinical practice. Systematic screening with AIMS is essential.
Pathophysiology
Dopamine Receptor Supersensitivity
Chronic D2 receptor blockade leads to compensatory upregulation and supersensitivity of postsynaptic D2 receptors in the dorsal striatum. When the antipsychotic is reduced or withdrawn, supersensitive receptors produce an exaggerated dopaminergic response, generating involuntary movements. This model explains why TD can be initially masked by increasing the DRBA dose and why dopamine-depleting agents (VMAT2 inhibitors) improve symptoms. However, this theory alone is insufficient: not all patients develop TD despite universal receptor upregulation, receptor changes normalize within weeks while TD persists for years, and animal models show inconsistent correlation between D2 density and dyskinetic behavior.
Oxidative Stress and Excitotoxicity
Compensatory increases in dopamine turnover during chronic blockade generate reactive oxygen species (ROS), hydrogen peroxide, and quinone metabolites via MAO metabolism. These cause lipid peroxidation, mitochondrial damage, and neuronal death in vulnerable striatal and nigral structures. Evidence includes elevated oxidative damage markers (malondialdehyde, 8-hydroxy-2-deoxyguanosine) in TD patients and associations between MnSOD gene polymorphisms and TD susceptibility.
GABAergic Neuronal Degeneration
Oxidative damage and direct drug toxicity cause degeneration of striatal GABAergic interneurons, impairing GABAergic inhibition within the direct and indirect basal ganglia motor pathways. This disinhibition of the thalamocortical motor circuits produces characteristic involuntary movements. Abnormal glutamatergic neurotransmission at corticostriatal synapses further contributes to excitotoxic damage.
Maladaptive Synaptic Plasticity
The integrative "maladaptive plasticity" hypothesis proposes that D2 receptor hypersensitivity combined with oxidative stress-induced degeneration of striatal GABAergic neurons creates aberrant long-term potentiation (LTP) and long-term depression (LTD) at corticostriatal synapses. These maladaptive plastic changes produce "miscoded motor programs" that persist even after the offending drug is removed, explaining the chronic and often irreversible nature of TD.
Neuroimaging Findings
- Structural MRI: Brain imaging typically appears normal; quantitative studies show reduced caudate nucleus volumes in TD patients compared to non-TD patients and healthy controls
- PET imaging: Striatal D2 receptor binding studies yield mixed results; some show no significant difference between TD and non-TD patients, challenging the simple supersensitivity model
- Iron deposition: Differences in iron deposition-related T1/T2 relaxation times have been reported in basal ganglia of TD patients, potentially reflecting neurodegeneration
Risk Factors
Non-Modifiable Risk Factors
| Risk Factor | Details |
|---|---|
| Age >55 years | Strongest risk factor; 5x higher incidence; strong linear correlation between ages 40–70 |
| Female sex | Higher risk overall, especially postmenopausal (estrogen may have neuroprotective effects) |
| African American descent | Higher prevalence; pharmacogenetic and pharmacokinetic factors |
| Mood disorders | Bipolar/MDD patients have higher vulnerability than schizophrenia patients, even at lower doses |
| Pre-existing brain injury or cognitive impairment | Decreased functional reserve increases susceptibility |
| Intellectual disability | TD prevalence 42–45% in institutionalized patients on antipsychotics |
Modifiable Risk Factors
| Risk Factor | Details |
|---|---|
| FGA use (vs. SGA) | Greatest risk; tight D2 binding |
| Duration of DRBA exposure | Cumulative dose-dependent; longer exposure = higher risk |
| Higher DRBA dose | Dose-dependent relationship |
| Concurrent anticholinergic use | Chronic anticholinergic use may increase TD risk |
| Early extrapyramidal symptoms | Drug-induced parkinsonism or akathisia predicts higher subsequent TD risk |
| Diabetes mellitus | Independent modifiable risk factor |
Genetic Risk Factors
| Gene / Polymorphism | Role | Evidence |
|---|---|---|
| DRD3 Ser9Gly | Dopamine D3 receptor variant | Most replicated association; OR 1.33 (pooled meta-analysis); OR 2.5–2.8 for severe TD in Gly carriers |
| CYP2D6 | Cytochrome P450 metabolizer status | Poor metabolizer alleles increase risk: OR 1.43 (95% CI 1.06–1.93); impaired metabolism leads to higher drug exposure |
| HSPG2 (rs2445142) | Heparan sulfate proteoglycan 2 (blood-brain barrier component) | Top GWAS finding; G allele overrepresented in TD; associated with higher HSPG2 expression in postmortem cortex |
| MnSOD/SOD2 (Ala-9Val) | Manganese superoxide dismutase (antioxidant enzyme) | -9Ala (high activity) allele may protect; Val9 associated with risk in some meta-analyses |
| DRD2 | Dopamine D2 receptor variants | OR 1.30–1.80 |
| HTR2A / HTR2C | Serotonin receptor variants | Associated in some studies |
Clinical Features
Classic Orobuccolingual TD
The hallmark presentation involves repetitive, stereotyped, purposeless movements of the mouth, tongue, and jaw:
- Tongue: Protrusion ("fly-catching"), lateral movements ("bon-bon sign"), rolling within the mouth, pressing against the cheek
- Lips: Lip smacking, lip pursing, lip puckering, fish-mouth movements
- Jaw: Chewing movements, lateral jaw deviations, bruxism
- Face: Grimacing, forehead furrowing, blepharospasm
Key features: movements are partially suppressible by volition, diminish during purposeful oral activity (speaking, eating), worsen with distraction and emotional stress, and disappear during sleep.
Distinguishing TD from HD Chorea
In patients with orobuccolingual dyskinesias, forehead chorea can help distinguish HD from TD. Forehead chorea is common in HD but unusual in isolated TD. However, clinicians should be aware that TD can co-occur in patients with HD who are taking dopamine-blocking medications (Continuum 2025).
Limb TD
- Upper extremities: choreiform finger movements ("piano-playing" fingers), wrist flexion-extension, arm flinging
- Lower extremities: foot tapping, toe movements, ankle flexion-extension, restless leg-like movements
- Impaired gait, balance, and fine motor function
Truncal and Respiratory TD
- Trunk: Body rocking, pelvic thrusting, axial dystonia, lateral trunk flexion
- Respiratory dyskinesia: Fast, irregular breathing with gasping, sighing, grunting, and forceful breathing; laryngeal endoscopy shows intermittent partial glottic obstruction; often misdiagnosed as asthma, COPD, or anxiety
- Laryngeal involvement: Can cause speech disorders and stridor; potentially life-threatening in severe cases
Coexisting Tardive Phenomena
A study of 100 patients with tardive syndromes found that 35% had two or more coexisting tardive phenomena:
| Tardive Subtype | Frequency |
|---|---|
| Orobuccolingual dyskinesia (classic TD) | 72% |
| Tardive tremor | 30% |
| Tardive akathisia | 22% |
| Tardive dystonia | 16% |
| Tardive tics (tourettism) | 4% |
| Tardive myoclonus | 1% |
Impact on Quality of Life
- Speech: Difficulty reported by 73% of patients
- Swallowing: Dysphagia with aspiration risk
- Dental: Damage from bruxism
- Social: Embarrassment, social withdrawal, stigma
- Functional: Impaired gait, balance, fine motor, and eating/swallowing
- Psychological: Depression, anxiety, reduced self-esteem
Differential Diagnosis
| Condition | Key Distinguishing Features |
|---|---|
| Huntington disease | Autosomal dominant family history; progressive dementia; generalized flowing chorea (not stereotyped); forehead chorea common; caudate atrophy on MRI; CAG repeat expansion in HTT |
| Spontaneous dyskinesia of schizophrenia | Reported in drug-naive patients (prevalence 0.5–15% in untreated schizophrenia); no temporal relationship to DRBA exposure |
| Edentulous dyskinesia | Orofacial movements in edentulous patients; improves with properly fitting dentures; no DRBA exposure required |
| Meige syndrome | Idiopathic cranial dystonia: blepharospasm + oromandibular dystonia; no DRBA exposure; dystonic (sustained) rather than choreiform; onset 5th–6th decade |
| Wilson disease | Young onset (<40); Kayser-Fleischer rings; low ceruloplasmin; elevated 24h urine copper; liver involvement; treatable with chelation |
| Autoimmune chorea | Acute/subacute onset; psychiatric features; autoantibody panel positive (anti-NMDAR, SLE, etc.) |
| Tourette syndrome | Childhood onset (<18 years); premonitory urge; waxing-waning course; vocal tics; suppressible |
| Functional (psychogenic) dyskinesia | Distractibility, entrainment, inconsistency; acute onset; may remit with suggestion |
Key clinical features that help confirm TD: Temporal relationship to DRBA exposure (minimum 3 months, or 1 month if age >60); characteristic orobuccolingual stereotypic pattern; amelioration by purposeful action; augmentation by distraction; partial volitional suppressibility; absence during sleep; history of improvement with DRBA dose increase (masking).
Assessment
Abnormal Involuntary Movement Scale (AIMS)
The AIMS is the standard assessment instrument: a 12-item clinician-rated scale administered in approximately 10 minutes.
| Items | Body Region / Domain |
|---|---|
| 1–4 | Facial and oral: (1) muscles of facial expression, (2) lips and perioral area, (3) jaw, (4) tongue |
| 5–6 | Extremity movements: (5) upper extremities, (6) lower extremities |
| 7 | Trunk movements |
| 8 | Global severity of abnormal movements |
| 9 | Incapacitation due to movements |
| 10 | Patient awareness of movements |
| 11–12 | Dental status (current dental problems; dentures present) |
Scoring (Items 1–7): 0 = none; 1 = minimal (low amplitude, present during some but not most of exam); 2 = mild (low amplitude most of exam, or moderate amplitude some); 3 = moderate (moderate amplitude most of exam); 4 = severe (maximal amplitude most of exam).
AIMS Examination Procedure
- Patient seated, hands on knees, legs slightly apart, feet flat on floor
- Observe at rest with hands hanging unsupported
- Ask patient to open mouth and protrude tongue (twice)
- Ask patient to tap thumb against each finger rapidly for 10–15 seconds (each hand) — activating movements reveal latent dyskinesia
- Flex and extend each arm
- Stand up and walk a few paces, turn, walk back
- Observe all body areas for involuntary movements throughout
Schooler-Kane research criteria for TD: Score of ≥2 (mild) in at least 2 body regions or ≥3 (moderate/severe) in at least 1 region, on two separate examinations at least 3 months apart.
Other Assessment Tools
- DISCUS (Dyskinesia Identification System: Condensed User Scale): 15-item scale; total score ≥5 should prompt further clinical evaluation; 96% agreement with AIMS-defined TD
- ESRS (Extrapyramidal Symptom Rating Scale): Broader scale assessing TD and other drug-induced movement disorders
- Patient Global Impression of Change (PGIC): Patient-reported outcome used in clinical trials
Prevention
Prevention and Monitoring Framework
- Before initiating any DRBA: Document clear indication, obtain informed consent regarding TD risk, perform baseline AIMS
- Use lowest effective dose for shortest duration — the single most important prevention strategy
- Prefer SGAs over FGAs when clinically appropriate; within SGAs, prefer those with lower D2 affinity (clozapine > quetiapine > others)
- Avoid chronic anticholinergic co-prescribing when possible (increases TD risk)
- Monitor regularly:
- Every 3 months for patients on FGAs
- Every 6 months for high-risk patients (elderly, FGA use, prior EPS, mood disorders, diabetes, female sex)
- Every 12 months for patients on SGAs considered lower risk
- If TD is detected: Immediately reassess risk-benefit; consider dose reduction, switch (to clozapine or quetiapine), or discontinuation
- Distinguish withdrawal-emergent dyskinesia (self-limited, resolves in 4–12 weeks) from covert/masked TD (persists >3 months after discontinuation, indicating true TD)
Metoclopramide
Metoclopramide carries an FDA black box warning against use for >12 weeks due to TD risk. It is the most common non-psychiatric cause of TD. Prochlorperazine and promethazine also carry risk.
Treatment
VMAT2 Inhibitors (First-Line, FDA-Approved)
Mechanism of action: VMAT2 inhibitors block the vesicular monoamine transporter 2 protein on presynaptic vesicle membranes. VMAT2 normally transports cytoplasmic monoamines (dopamine, serotonin, norepinephrine) into synaptic vesicles for storage and release. By blocking this transporter, dopamine remains in the cytoplasm where it is rapidly degraded by MAO, causing dose-proportional depletion of presynaptic dopamine stores. Critically, VMAT2 inhibitors do not bind to postsynaptic dopamine receptors, avoiding further supersensitization.
Valbenazine (Ingrezza)
FDA approved April 2017 for TD; April 2023 for HD chorea.
| Parameter | Details |
|---|---|
| Starting dose | 40 mg once daily for 1 week |
| Maintenance dose | 80 mg once daily |
| Half-life | 15–22 hours |
| CYP2D6 poor metabolizers | Maximum 40 mg/day |
| Hepatic impairment | Mild-moderate: 40 mg max; severe: avoid |
| Renal impairment | Severe: not recommended |
| Key side effects | Somnolence (most common), fatigue, headache, dry mouth, akathisia (~10%) |
| QTc effect | Mean change 1.1–2.1 ms (not clinically significant) |
| Black box warning | Depression and suicidality (from HD data); screen for depression before initiating |
KINECT trial program:
| Trial | Design | N | Duration | Key Results |
|---|---|---|---|---|
| KINECT 3 | Phase 3 RCT | 234 | 6 weeks | 80 mg: AIMS −3.2 vs. −0.1 placebo (p<0.001); 40 mg: −1.9 |
| KINECT 3 Extension | Open-label | 198 | 42 weeks | 40 mg: −3.0; 80 mg: −4.8; AIMS returned toward baseline after 4-week washout |
| KINECT 4 | Phase 3, open-label | ~160 | 48 weeks | >80% had ≥50% AIMS improvement; 90% rated "much/very much improved" at week 48 |
| KINECT-HD | Phase 3 RCT | — | — | Approved for HD chorea (2023) |
Deutetrabenazine (Austedo)
FDA approved August 2017 for TD and HD chorea. Deuterated form of tetrabenazine with prolonged half-life and improved side effect profile.
| Parameter | Details |
|---|---|
| Starting dose | 12 mg/day (6 mg twice daily) |
| Titration | Increase by 6 mg/day weekly |
| Maximum dose | 48 mg/day (36 mg in CYP2D6 poor metabolizers or with strong CYP2D6 inhibitors) |
| Half-life | 9–10 hours (also available as extended-release formulation — Austedo XR) |
| Key side effects | Somnolence, headache, anxiety, depression, dry mouth, akathisia, QTc prolongation (mean 4.5 ms at 24 mg) |
| Contraindications | Hepatic impairment; uncontrolled depression or suicidal ideation; concurrent MAOIs |
| Black box warning | Depression and suicidality (from HD data) |
Key trials:
| Trial | Design | N | Duration | Key Results |
|---|---|---|---|---|
| ARM-TD | Phase 2/3 RCT, flexible-dose | 117 | 12 weeks | AIMS −3.0 vs. −1.6 placebo (p=0.019) |
| AIM-TD | Phase 3 RCT, fixed-dose | 298 | 12 weeks | 24 mg: −3.2 (p=0.003); 36 mg: −3.3 (p=0.001); 12 mg: −2.1 (NS) |
Tetrabenazine (Xenazine) — Off-Label for TD
FDA approved in 2008 for HD chorea only (not TD). Administered 3 times daily; half-life 5–7 hours. Maximum daily dose for HD: 50 mg; for TD: 75 mg (up to 200 mg in some studies). More depression (19%), insomnia (22%), and akathisia (19%) compared with valbenazine/deutetrabenazine. Metabolized by CYP2D6 — higher doses required in extensive metabolizers. Largely supplanted by the newer VMAT2 inhibitors due to less favorable dosing and side effect profile.
All VMAT2 Inhibitors
MAOIs are contraindicated with all three VMAT2 inhibitors. Screen for depression and suicidal ideation before initiation (all carry boxed warnings from HD data). Valbenazine does not appear to cause depression as frequently as tetrabenazine or deutetrabenazine. CYP2D6 genotyping should be considered — poor metabolizers require dose adjustment for all three agents.
Second-Line and Adjunctive Agents
| Agent | AAN Evidence Level | Dose | Key Evidence |
|---|---|---|---|
| Clonazepam | Level B (probably improves) | 0.5–4 mg/day | GABAergic enhancement; better for dystonic than choreiform TD; limited by sedation, tolerance, dependence |
| Ginkgo biloba (EGb-761) | Level B (probably improves) | 240 mg/day | Largest RCT: 51% response vs. 5% placebo (N=157, 12 weeks); antioxidant mechanism; well tolerated |
| Amantadine | Level C (might be considered) | 100–300 mg/day | NMDA glutamate antagonist; small magnitude of effect for TD specifically |
| Vitamin B6 (pyridoxine) | Grade B (Canadian guidelines) | Up to 1,000 mg/day | Short-term use; avoid neuropathy at high doses |
| Propranolol | Limited | 20–80 mg/day | 64% improvement in one study of 47 patients; may be more useful for tardive akathisia |
| Clozapine | — | Variable | Lowest TD risk of all antipsychotics; may reverse TD; recommended when ongoing antipsychotic is needed |
Agents with Negative or Mixed Evidence
- Anticholinergics (benztropine, trihexyphenidyl): May worsen classic orobuccolingual TD (though they help tardive dystonia — see separate article). Chronic anticholinergic use is itself a risk factor for TD development
- Vitamin E: Large VA Cooperative Study #394 showed no efficacy for established TD; Cochrane review (2018): "evidence is weak and limited." May have modest early protective role but no benefit once TD is established
- Levetiracetam: Insufficient evidence to support or refute use
Deep Brain Stimulation for Refractory TD
For severe, disabling TD refractory to maximal pharmacological treatment lasting >1 year.
- Target: Bilateral globus pallidus internus (GPi), posteroventrolateral portion
- Outcomes: Efficacy ranges from 28% to 100% reduction in AIMS/BFMDRS scores; largest cohort (19 patients): 63% AIMS improvement over 12–132 months
- STN-DBS: 88% BFMDRS and 94% AIMS improvement over 12–105 months (10 patients)
- Onset of benefit: Immediate to months; sustained for years
- Safety: Overall side effect rate ~9%; transient mood changes, speech impairment (~30% with bilateral stimulation)
Treatment Algorithm
Stepwise Management of TD
- Step 1 — Prevention: Baseline AIMS; use lowest effective dose of DRBA; prefer SGAs (especially clozapine/quetiapine); regular monitoring
- Step 2 — Address the DRBA: Taper/discontinue if psychiatrically feasible (coordinate with psychiatry); if needed, switch to clozapine or quetiapine
- Step 3 — VMAT2 inhibitors (first-line): Valbenazine 40→80 mg/day OR deutetrabenazine 12→48 mg/day; titrate over 1–4 weeks; assess response at 6–12 weeks
- Step 4 — Adjunctive therapy: Add clonazepam 0.5–4 mg/day, EGb-761 240 mg/day, or amantadine if partial response to VMAT2 inhibitor
- Step 5 — Refractory TD: Consider GPi-DBS for severe, disabling TD unresponsive to maximal pharmacotherapy for ≥1 year
Special Populations
Elderly (>60 Years)
- Highest risk population (5x higher incidence); up to 30% develop TD within 1 year of exposure
- Spontaneous remission rate inversely correlated with age (<60: 3x more likely to improve)
- DSM-5 permits TD diagnosis after only 1 month of DRBA exposure (vs. 3 months for younger adults)
- Greater susceptibility to neurodegenerative changes from oxidative stress; lower monitoring threshold (AIMS every 3–6 months)
Children and Adolescents
- TD point prevalence 5–20% in antipsychotic-treated youth; likely underdiagnosed (~1% from clinical databases)
- Withdrawal-emergent dyskinesia (self-limiting) is more common in children than persistent TD
- VMAT2 inhibitors are not FDA-approved for pediatric use
- Growing concern due to increasing off-label antipsychotic prescribing (autism, ADHD, conduct disorder)
Intellectual Disability
- TD prevalence 42–45% in institutionalized patients on antipsychotics — substantially higher than general population
- Diagnostic challenges: communication barriers, movements attributed to underlying condition, incomplete medication histories
- Clinical trials establishing VMAT2 inhibitor safety/efficacy excluded individuals with ID
- Case series: marked improvement with valbenazine 40–80 mg daily within 2–3 weeks (speech, mobility, ADL independence)
Emerging Developments (2023–2025)
- NBI-1065890 (Neurocrine Biosciences): Next-generation selective VMAT2 inhibitor; Phase 2 RCT initiated January 2026
- Austedo XR: Extended-release deutetrabenazine formulation allowing once-daily dosing
- Pharmacogenomic-guided prescribing: CYP2D6 testing before VMAT2 inhibitor initiation gaining traction
- Digital biomarkers: Remote TD monitoring technologies under investigation
- IMPACT-TD Registry (2024): Largest study on holistic effects of TD; only 36% of patients with psychotic disorders received TD diagnoses (vs. 50% with mood disorders), highlighting significant underdiagnosis in psychotic illness
References
- Continuum (Minneap Minn). August 2025; 31(4 Movement Disorders). Huntington Disease and Chorea (pp 1066–1087).
- Carbon M, Hsieh CH, Kane JM, Correll CU. Tardive dyskinesia prevalence in the period of second-generation antipsychotic use: a meta-analysis. J Clin Psychiatry. 2017;78(3):e264–e278.
- Hauser RA, Factor SA, Marder SR, et al. KINECT 3: a phase 3 randomized, double-blind, placebo-controlled trial of valbenazine for tardive dyskinesia. Am J Psychiatry. 2017;174(5):476–484.
- Fernandez HH, Factor SA, Hauser RA, et al. Randomized controlled trial of deutetrabenazine for tardive dyskinesia: the ARM-TD study. Neurology. 2017;88(21):2003–2010.
- Anderson KE, Stamler D, Davis MD, et al. Deutetrabenazine for treatment of involuntary movements in patients with tardive dyskinesia (AIM-TD). Lancet Psychiatry. 2017;4(8):595–604.
- Bhidayasiri R, Fahn S, Weiner WJ, et al. Evidence-based guideline: treatment of tardive syndromes. Neurology. 2013;81(5):463–469.
- Ricciardi L, Pringsheim T, Barnes TRE, et al. Treatment recommendations for tardive dyskinesia. Can J Psychiatry. 2019;64(6):388–399.
- Factor SA. Management of tardive syndrome: medications and surgical treatments. Neurotherapeutics. 2020;17(4):1694–1712.
- Frank S, Anderson KE, Fernandez HH, et al. Safety of deutetrabenazine for the treatment of tardive dyskinesia and chorea associated with Huntington disease. Neurol Ther. 2024;13(3):655–675.
