Clinical manifestations

Presentation and course

Akathisia and acute dystonic reactions may develop within minutes of drug intake (40). More commonly, acute dystonia occurs before akathisia, generally within hours to days of drug initiation or dose increase. Ninety-five percent of acute dystonic reactions occur within 4 days of drug initiation (06).

Acute dystonic reactions are involuntary sustained muscle contractions causing an abnormal posture that usually lasts for about 20 to 30 minutes at a time and is often painful; the contractions remit and recur a variable number of times before resolving entirely (33; 13). The cranial and cervical muscles are most commonly involved. The head may be turned, flexed, or extended. Jaw, tongue, and muscles of facial expression may be involved in isolation or together. Patients with tongue involvement frequently describe it as feeling swollen, most likely the subjective sensation of involuntary tongue contractions and not the result of actual swelling (personal observation). Oculogyric crisis (described previously only with postencephalitic parkinsonism), in which the eyes deviate conjugately up and to one side, may occur in concert with other dystonic reactions or by itself. Pisa syndrome is an uncommon form of dystonia that causes truncal flexion towards either side (83). Pure truncal flexion is called camptocormia. The Pisa syndrome may occur as a tardive syndrome or as an acute dystonic reaction. It has been observed in Parkinson disease and has been associated with several non-neuroleptic medications.

Akathisia is more difficult to recognize (86). It is an inability to remain still due to a sense of inner restlessness (02). People are gripped by a compulsion to move, as if a "foreign force" (45) has taken over, making them unable to remain in one place. When seated, the patient will rock in place or change positions, often standing every few seconds, and will appear fidgety. While standing, the patient will typically march in place or shift weight from side to side (11). If observed outside of a formal evaluation, akathisic patients move continuously and find it difficult to watch television, engage in conversations, or sleep because of difficulty remaining still. When mild, it may be perceived by the patient or observer as anxiety (45; 29). It may be impossible to distinguish akathisia from psychosis when the patient is uncommunicative due to psychosis (85; 86) or dementia. Akathisia may persist for long periods if neuroleptics are continued (56). Akathisia is an uncomfortable sensation and, when severe, patients may consider it a greater problem than the psychosis being treated (personal observation). As a result, it is associated with higher measures of suicidality than similarly treated patients without this problem. Acute movement disorders induced by antiemetics and antipsychotics, when not used for psychosis, are less likely to be recognized than when antipsychotics are used for schizophrenia (89).

Clozapine may cause asterixis, presumably even causing falls due to knees buckling, but how quickly this may develop is unknown (30).

Drugs of abuse may also cause acute movement disorders. Cocaine is known to cause chorea and dystonia via dopamine reuptake inhibition (20). Amphetamine use can result in punding, which is a repetitive purposeless stereotyped behavior such as excessive hair brushing (26).

Prognosis and complications

Acute dystonic reactions consistently resolve when the offending drug is stopped, and consistently respond to anticholinergics or diphenhydramine. After the acute episode resolves, patients should continue taking the anti-dystonia medication orally for the next couple of days to prevent recurrence, but data on treatment are lacking.

Akathisia has a more complex prognosis. In most cases, akathisia resolves if the medication is stopped, but it may take longer. It may persist if the neuroleptic is continued, unlike acute dystonia, which is self-limited. Akathisia can even evolve into a tardive syndrome. Although akathisia is believed to be a common reason for medication noncompliance among the psychotic, it is not known if akathisia precludes a good antipsychotic response. When akathisia is recognized, the prognosis is good. The akathisia is treated directly or the offending neuroleptic is stopped, and (if possible) a new one of a different chemical family and of lower potency is started. Prognosis is obviously much worse for patients whose akathisia is unrecognized or who are treated in institutions against their wills, or with depot injections. Interestingly, akathisia may be prognostic of a poor antipsychotic response (53).

Clinical vignette

Vignette 1. An 80-year-old man, without any known neurologic problems, was admitted to the hospital for evaluation of severe abdominal pain. On his first night he became confused and tried to climb out of the bed. He was given an intramuscular injection of haloperidol 1 mg. A few hours later he became mildly agitated, climbed out of bed again, and was given another injection of the haloperidol. Over the next 3 days this continued while his abdominal problem, undiagnosed, resolved. He was discharged from the hospital on haloperidol 2 mg daily, which no one stopped. He was evaluated 3 months later and was severely parkinsonian and akathisic. He apologized to me repeatedly as he stood up from his chair every minute or so, explaining that he was unable to sit still. His haloperidol was stopped, and his akathisia resolved within 3 months. At that time, his parkinsonism improved but was still present.

Vignette 2. When asked if he had seen much akathisia when he used an intravenous cocktail that included metoclopramide, one neurologist/headache specialist responded that he had never observed it. He noted, however, that occasionally patients became extremely anxious and refused to remain in the clinic for the usual observation period; these patients had unrecognized akathisia (44).

Biological basis

Etiology and pathogenesis

By definition, these syndromes are drug induced. Most commonly, they are caused by dopamine receptor blocking drugs. Dopamine depleting drugs such as alpha-methylparatyrosine, used to treat hypertension, which block the synthesis of catecholamines, can also cause acute dystonia (59). Akathisia has been well described as an adverse effect of the selective serotonin reuptake inhibiting antidepressants, yet these drugs induce parkinsonism or acute dystonia only rarely (35; 52). In a large retrospective chart review, nine patients of 1875 taking citalopram developed acute dystonic reactions (63). Certain antimalarials also have been reported to rarely induce dystonia, as does the antiepileptic, tiagabine (92). One case of bupropion-induced acute dystonic reaction has also been documented (19), which is surprising because the drug has mild dopaminergic properties. Levodopa frequently causes recurrent dystonia in Parkinson disease patients, but these usually involve a foot, are also associated with choreoathetoid movements, and resolve when levodopa is discontinued. Wilson disease may cause acute dystonia in children (23), and a variety of other disorders may very rarely cause acute dystonic spells in adults or children (08). Akathisia has been reported to affect about 7% of people treated with tetrabenazine for hyperkinetic movement disorders (46). It also has been reported to cause acute dystonia (12).

Only limited data support the theoretical explanations for either acute dystonia or akathisia. Acute dystonia can be induced in animals with injections directly into the brain using cholinergic agonists (14) or sigma opiate receptor blockers (87). All theories are based on the clinical observation that dopaminergic receptor blockade induces both acute dystonia and akathisia, but dopamine depleting drugs do as well. Two major theories to explain acute dystonia center on a dopamine imbalance: either too little or too much (55; 56; 13), or possibly variant dopamine receptors (49). Neuroleptics induce compensatory dopamine release, which presumably overstimulates the unblocked receptors in a manner similar to excess levodopa causing dystonia in Parkinson disease. A third theory holds that an acute insufficiency of dopamine stimulation causes dystonia, explaining the relationship to high doses and high potency neuroleptic drugs, the effectiveness of anticholinergics in treating the syndrome, and the early onset of the dystonia. A puzzling feature of the atypical antipsychotics has been their relative freedom from extrapyramidal side effects despite their antagonism for D2 receptors. One hypothesis for this feature, which may also be a contributing explanation for the development of acute dystonia, is the speed of dissociation of these drugs from the D2 receptor (41). Thus, rates of dissociation may have something to do with parkinsonism and the more acute side effects. However, aripiprazole, a partial dopamine D2 agonist, has been implicated in causing acute dystonic reactions in several case reports (94; 65). An interesting observation that complicates our understanding is that haloperidol, when given intravenously, is less likely to cause either of these problems than when given orally, even at high doses (60); however, IM haloperidol causes more acute movement disorders than oral haloperidol. Yet a large prospective study of intravenous metoclopramide given for the treatment of headache reported akathisia, but that was reduced by slowing the infusion from 2 to 15 minutes (74). Various studies have suggested a predisposition for the development of akathisia with low serum iron or ferritin, but other published work does not corroborate this notion (38).

A single PET study has shown a correlation between high binding of D1 and D2 specific dopamine receptor binding ligands in the putamen in normal and schizophrenic subjects with akathisia (25). An alternative theory suggests that akathisia may in part be the result of acute mesocortical dopamine insufficiency (35). How the selective serotonin reuptake inhibitor antidepressants cause akathisia is unknown. There may be subtypes of depression that are more likely to be associated with such a response (50), and there may be differences among the selective serotonin reuptake inhibitors and their affinity for the 5HT2C receptor. The effect may be the result of an indirect effect on dopamine.

There are no compelling theories to explain either akathisia or dystonia that integrate data regarding responses to treatment with medications of widely different types, the epidemiology of the syndromes, and the occurrence of both syndromes in postencephalitic parkinsonism.

Three cases of recurrent oculogyric crises were reported to occur despite withdrawal from the offending dopamine-blocking drug (77), blurring the distinction between acute and tardive dystonia (27).

Epidemiology

Acute dystonia is more common in males and the young in general, with peak incidence under 15 years of age (48). It decreases with age (03), becoming rare after 40 years of age. In a prospective study of 200 neuroleptic naive patients admitted to a psychiatric hospital, 45% developed acute dystonic reaction, without gender preference, on low doses of haloperidol (58). Another prospective study of 660 previously treated outpatients found acute dystonic reaction in only 8% of haloperidol-treated patients (84). In children, generalized dystonic reactions and oculogyric crises predominate, whereas adults tend to suffer isolated neck and cranial dystonias. A report describing acute dystonic reactions occurring in three young brothers when given metoclopramide for gastroenteritis (79) raises the question of a genetic predisposition. Akathisia seems to have no age or gender predilections (51; 76). Both are more common with high-potency neuroleptics, high doses of drugs, and depot injections (56). Estimates for the occurrence of akathisia are variable, ranging from under 10% in early studies to 70% (01), with a figure of 20% generally accepted (13) depending on criteria used (76). A review of akathisia incidence associated with the newest second-generation antipsychotics in subjects of controlled clinical trials found that exposure to each of the drugs was significantly increased compared to those assigned to placebo, and these subjects had all been on neuroleptics shortly before enrolling in the trials (18). Akathisia usually develops over days to weeks, following a time course similar to, but somewhat faster than, drug-induced parkinsonism (56). Ninety percent of cases develop within 73 days of the beginning of treatment, but some develop within minutes, especially after intravenous administration of a neuroleptic (25). In a study of migraine treatment with intravenous metoclopramide in nonpsychiatric patients, akathisia developed in 33 of 356 patients, mostly women, within minutes (29). There was no relationship to dose. Onset of akathisia after 3 months of a stable neuroleptic dose may be considered tardive akathisia (75).

An interesting anesthesiology study reported on extrapyramidal side effects in 45,766 patients who were given a single dose of perphenazine (4 mg or 8 mg) to prevent postoperative nausea (37). Only five cases of akathisia and one of acute dystonic reaction were noted in this retrospective chart review, an event rate of 1.3 per 100,000 patients.

Abnormalities of iron stores in the brain has been suggested as a risk factor for developing akathisia, but data supporting this hypothesis are weak (78).

Even the newer second generation antipsychotics may cause acute akathisia (47).

Prevention

The best prophylaxis against both acute dystonic reactions and akathisia is to avoid the drugs that cause them. The atypical antipsychotics either do not cause these side effects or do so at a significantly lower rate than the previous generation of antipsychotic drugs. It is important to note that this “second generation” of antipsychotics has not been shown to be more effective than the previous generation, with the exception of clozapine, but there are data suggesting fewer acute extrapyramidal side effects such as acute dystonia and akathisia (15). Ziprasidone has been increasingly, but still uncommonly, reported to cause acute dystonic reactions (88; 95), as have the other atypicals except for quetiapine and clozapine. Akathisia has also been associated with aripiprazole and risperidone (42; 67).

Prophylaxis against acute dystonia with antiparkinson medication has been studied and found to be helpful (43; 80; 91; 10; 13; 03). Use of low-potency neuroleptics, which have prominent anticholinergic activity, also reduce the incidence. However, the anticholinergics have potent side effects including dry mouth, constipation, blurred vision, memory impairment, urinary retention, and sexual dysfunction. The utility of amantadine in prophylaxis of acute dystonia has not been studied. Prophylaxis probably should be reserved for those at most risk and discontinued after a few weeks, when the risk falls.

Although akathisia is common, high quality studies of prophylaxis are lacking. A Cochrane review found no reliable data on this (73), and a metaanalysis found a single placebo controlled trial reporting benefit for diphenhydramine for akathisia against metoclopramide (62).

The suggestion that giving a neuroleptic intravenously rather than orally might reduce the likelihood of extrapyramidal side effects was not borne out in a study in which over 40% of migraineurs who received intravenous haloperidol experienced uncomfortable restlessness (60; 31).

Differential diagnosis

Acute dystonic reactions are easy to recognize. Unfortunately, factitious cases, involving medicated patients familiar with the syndrome, can be impossible to distinguish from real cases. These patients mimic the disorder for secondary gain. Naturally occurring dystonic syndromes such as torticollis, writer's cramp, and blepharospasm can look like dystonic reactions but are slower in onset, occur typically in nonpsychiatric patients, involve no precipitating medications, and do not resolve.

Tonic focal seizures, paroxysmal dystonias, and muscle spasm due to bony injury or subluxation (especially in the cases of atlantoaxial subluxation simulating torticollis or a subluxated jaw looking like jaw dystonia) can look like acute dystonia but should be easily distinguishable based on history and duration of the episodes. Hypocalcemia with carpopedal spasm may also need to be considered, as well as cephalic tetanus and drug intoxication (39). An interesting chart review found that acute dystonic reactions in children were misdiagnosed in the majority of cases (22; 93). The initial diagnoses had been conversion disorder, meningitis, epileptic seizure, mandibular subluxation, tetanus, and unknown brain pathology (22).

Akathisia may be confused with mania, stimulant drug intoxication, drug withdrawal, or an agitated emotional state (86; 51). One must be wary of missing akathisia, as it may be the cause for a paradoxical worsening of behavior in response to antipsychotic treatment (85; 86; 07). Equally important is the consideration of akathisia as a driving force for agitation in demented patients who receive antipsychotic drugs for psychiatric aspects of their dementias or for agitated behavior in children and adolescents (32). A summary of an international meeting on agitation noted that the majority or participants considered antipsychotics their first-line drug treatment when the etiology of the agitation was unknown (57). Therefore, worsening of agitation should be a warning sign. If the response to an antipsychotic is worsened behavior, akathisia must be considered. Because the patients are often unable to communicate, a high index of suspicion must be maintained. Restless legs (Ekbom syndrome) is not associated with dopamine blockade although it is relieved by L-DOPA and dopamine agonists. Patients develop uncomfortable sensations in their legs that are relieved by walking (51). These sensations occur primarily at night and interfere with falling asleep but do not occur to a significant degree during the day. This syndrome has been associated with iron deficiency in some cases. Finally, tardive dyskinesia or "pseudo akathisia" can cause a constellation of fidgety looking choreic movement in which patients look restless but are not (11).

Diagnostic workup

Usually no laboratory evaluation is required for either syndrome. Obviously when doubt exists about a diagnosis, the appropriate studies should be performed. Persistent neck deviation may require plain x-rays or a cervical CT scan to exclude atlantoaxial subluxation. An EEG may help evaluate a possible tonic seizure disorder. Toxicology screens are indicated when a patient has an apparent acute dystonic reaction but denies medication intake, or when a patient becomes agitated and drug abuse is considered. Phencyclidine has been reported to cause dystonia unresponsive to diphenhydramine (66). Drug withdrawal is clearly more difficult to measure but traces of some drugs, such as barbiturates and benzodiazepines, which have long serum half-lives, may still be present during withdrawal.

Imaging studies are rarely useful but are indicated when signs are focal, when a medication history is lacking, and when infection is considered a possibility. When AIDS is a possible confounding diagnosis, the threshold for imaging and for lumbar puncture should fall significantly. For nonimmunocompromised patients, spinal fluid analysis is needed only if signs suggesting meningitis or encephalitis are present (fever, stiff neck, obtundation, unexplained behavioral changes atypical for psychosis).

Management

Drug treatment. Acute dystonia is easily treated with intramuscular or intravenous benztropine, diphenhydramine or diazepam, or resolves by itself whether the precipitating drug is discontinued or not. Prophylaxis with anticholinergics or diphenhydramine lowers the risk but increases adverse effects. When prophylaxis is used it can be reduced after 1 to 2 weeks.

Akathisia treatment is less straightforward. Several double-blind studies have demonstrated the benefits of low dose beta-blockers. Propranolol, a beta 1 and beta 2 lipophilic antagonist, has been the most studied and has been compared to betaxolol, a central beta 1 antagonist (21) as well as nadolol, atenolol, sotalol, metoprolol and pindolol (02). Unfortunately the number of patients involved in these studies has been small, leading to contradictory results. Propranolol appears to work as well or better than any other beta-blocker at doses as low as 10 to 20 mg three times daily. There is no answer yet as to whether the beta 1 or beta 2 receptor is more important. Selective blockers of each have been found to alleviate akathisia, once the blood-brain barrier is crossed (28).

Anticholinergics have long been used for akathisia and appear to work best in cases associated with parkinsonism (11). A Cochrane Review (54) found “no reliable evidence to support or refute the use of anticholinergics for people suffering from akathisia.” Benzodiazepines, clonidine, an alpha 2 partial agonist, amantadine, ritanserin, and mianserin (69), both 5HT2 antagonists, as well as other drugs have also been reported to improve akathisia. One report cites cyproheptadine as a potential therapy (04). It should be noted that some review articles on treating akathisia fail to distinguish between acute and tardive akathisia, which have very different treatments (72).

An open-label study described a beneficial effect of nicotine patches in neuroleptic-induced akathisia (05). If confirmed, this result would potentially explain the high cigarette use among psychiatric in-patients but would add a cholinergic agonist (nicotine) to the cholinergic (muscarinic) antagonists as a potential therapy. Another open-label report describes a positive, but brief (3-day), trial of the 5-HT 1D receptor agonist, zolmitriptan, 2.5 mg three times daily (34).

A prospective randomized controlled trial reported that midazolam, a rapid-acting benzodiazepine, reduced akathisia, within 5 minutes, significantly faster than diphenhydramine, although with greater sedation (64). Another report requiring duplication found benefit from trazodone 100 mg every night in a double blind, placebo controlled, crossover trial involving only 13 subjects (82). Mirtazapine (15 mg) has also been proposed, with only limited data (68).

A meta-analysis supported the use of mirtazapine, an antidepressant anxiolytic that increases norepinephrine secretion while blocking 5HT2A receptors, for acute akathisia (70). However, this was based on only two double-blind placebo-controlled trials in which 15 mg was given in the morning. Only 20% of the patients went into complete remission, and the drug took about 3 days to work.

A consensus of experts recommended that for pediatric cases of acute dystonic reactions, diphenhydramine and anticholinergics are the best first choices, and that anticholinergics, propranolol, clonazepam, and mirtazapine are the first-line choices for acute akathisia (71).

The best approach depends on the situation. If the psychosis is improving, the akathisia should be treated with anticholinergics if parkinsonism is present. If parkinsonism is not present, propranolol should be used. If anxiety is complicating the picture, a benzodiazepine is recommended. If psychosis persists and akathisia is present, then use of a neuroleptic of a different family and lower potency, or the dopamine-depleting drug tetrabenazine, if possible, is indicated (40). Other “second line” drugs that have been advocated include cyproheptadine, mirtazapine, trazodone, ritanserin, mianserin (34).

Special considerations

Pregnancy

When the patient is pregnant, especially in the early stages, the fewer drugs used the better. Thus, pregnant patients should be reassured and alterations of the neuroleptic should be attempted before additional drugs are used to suppress the neuroleptic's adverse effects.

Anesthesia

Anesthesia is not a problem for patients on neuroleptics, except that droperidol, a commonly used dopamine receptor blocker antiemetic and induction agent, can precipitate an acute dystonic reaction or akathisia. Thus, a patient on neuroleptic may be more at risk for these problems when droperidol is used and may awaken with postoperative pain and an acute dystonic reaction, or a severe need to move around.