History and Phenomenology

“Akathisie” was originally coined by the Czechoslovakian neuropsychiatrist Ladislav Hascovec shortly after the turn of the 20th century,Reference Berrios^1–Reference Sachdev³ probably in 1901. He created the term in Czech from the Greek α (not) + κάθιςω (to be made to sit down) — ie, inability to sit. Hascovec considered akathisia (as the term later came to be spelled) to be primarily a movement disorder (but one with a more psychiatric etiology, such as a hysterical or conversion disorder), and only 1 of the 2 patients he initially described appeared to have the characteristic accompanying subjective discomfort and “need to move” that we now associate with the condition. In 1923, akathisia was described in patients with post-encephalitic parkinsonism by Paul Robert Bing and Jean-Athanase Sicard, but again it was considered more of a psychiatric condition.Reference Sachdev³ After the discovery that antipsychotic medications frequently caused the disorder in the 1950s, it was later grouped with other movement problems that commonly occur with antipsychotics, including parkinsonism and dystonia, all of which came to be referred to as extrapyramidal side effects (EPSE) or extrapyramidal signs (EPS).Reference Sachdev³

Although the term “akathisia” is often pronounced as rhyming with dyskinesia (and sometimes even spelled akathesia),Reference Baden, Prodany, Wiener and Hoffman⁴ it derives from very different Greek roots (there is no “kinesis” or movement in the etymology of akathisia) and is more correctly spelled “akathisia,” and pronounced æ.kə.ˈθɪ.si.ə (using International Phonetic Alphabet symbols, where “ɪ” is pronounced as in “kit” or “bin” and “i” is pronounced as in “glorious”).

Akathisia straddles the boundary between being a symptom (ie, a subjective problem described by the patient) and a sign (an objective problem observed by the clinician). The subjective component can exist independently of the objective component, particularly when the condition is mild. The opposite, if there is an objective component but no subjective discomfort, has been called pseudoakathisia—a troublesome concept discussed below.

The subjective discomfort of akathisia is often extraordinarily difficult for the patient to describe, to some extent because there are few subjective states to which it can be compared. Hence, patients often use terms such as “anxiety” or “itching,” even though these do not really capture the essence of the condition. In addition, since many clinicians have never experienced it, there is often a lack of common ground in communicating the problem. Characteristic descriptions that two of us (JBL and MAS) have heard over the years include, “It feels like my legs are on fire inside,” “It makes me want to run around the room or leap out of my bed,” “It’s like a deep itching and tingling inside my bones and moving around relieves it a little,” “It’s sort of like that feeling you get after your leg falls asleep and then wakes up, but it doesn’t go away,” and “I want to jump out of my skin.” The severity of subjective discomfort ranges from “mildly annoying” (and easily relieved by moving a limb or shifting position) to “absolutely intolerable,” which is associated with extreme dysphoria and a sense of “impending doom.” The most severe cases have been linked to suicidal ideation, aggression, and violence.Reference Keckich^5–Reference Galynker and Nazarian⁹ Akathisia is an important cause of poor medication adherence.Reference Lipinski, Mallya, Zimmerman and Pope^10–Reference Van Putten¹² The presence of akathisia also exacerbates any psychiatric symptoms present,Reference Van Putten^12–Reference Lohr and Wisniewski¹⁶ often leading clinicians to increase inappropriately the offending agents, such as selective serotonin reuptake inhibitors (SSRIs) or antipsychotics.

The objective sign of akathisia is a movement disorder. When mild to moderate in severity, the lower extremities are predominantly involved, usually from hips to ankles, and the movements take the form of shifting positions while standing, and rocking or moving the feet around while sitting. The predilection to affect primarily the lower extremities is often helpful in differentiating akathisia from other antipsychotic-induced side effects that tend to affect other body regions (described more below). With increasing severity, however, akathisia can involve the entire body, resulting in near incessant writhing and rocking movements often accompanied by jumping around, running, and occasionally flinging oneself out of a chair or a bed.Reference Lohr and Browning¹⁷ Although we have never personally seen a case of akathisia that did not affect the lower extremities to some degree, there are reports of akathisia occurring primarily in other body regions (such as suboccipital muscles)Reference Hirose¹⁸ or in strange distributions, such a hemi-akathisia (affecting only one side of the body).Reference Yamashita, Horiguchi, Mizuno, Kuramoto, Yamawaki and Inami¹⁹ Strange or unusual distributions of either subjective or objective components of akathisia should increase concern that there may be another process present influencing the clinical presentation, such as an infarction or abscess.Reference Yamashita, Horiguchi, Mizuno, Kuramoto, Yamawaki and Inami¹⁹

Although commonly considered a type of movement disorder or EPS, in fact akathisia should be considered more a sensorimotor disorder because of the powerful sensory component, which is a defining characteristic of the condition. In fact, the sensory component may be the primary problem, with motor signs being secondary to the restlessness and need to move.

Potential Subtypes

Akathisia has been divided into several subcategories, mainly in relation to drug-induced forms of the condition. The differences between these subtypes are generally based on the time course of the condition, because the actual clinical presentations (save for pseudoakathisia) are indistinguishable from one another.

• ∙ Acute, subacute, and chronic akathisia:Reference Bratti, Kane and Marder^20, Reference Sachdev²¹ Akathisia usually occurs within a few days to weeks of initiating an antipsychotic medication or increasing the dose. During the initial weeks, it is considered acute, and later it is considered subacute or chronic. Akathisia may develop within days and maintain severity at that level over time, or it may gradually increase or decrease in severity over time. When increasing in severity, the subjective and objective components become more pronounced and spread to other body regions. Chronic akathisia simply refers to akathisia that has been present a long time, usually several months or more, and is a different concept from tardive akathisia, described below.

• ∙ Withdrawal akathisia:Reference Lang²² This condition, indistinguishable phenomenologically from acute akathisia, occurs upon dosage decrease or withdrawal of antipsychotic agents. Withdrawal akathisia usually appears within about 2 weeks of discontinuation, and disappears within about 6 weeks. If it lasts longer, then it probably represents tardive akathisia, as described below.

• ∙ Tardive akathisia:Reference Bhidayasiri and Boonyawairoj^23, Reference Nishikawa, Koga, Uchida and Tanaka²⁴ Again, tardive akathisia is indistinguishable in clinical appearance from acute akathisia, but its time of onset and course resembles those of tardive dyskinesia. In other words, it (1) occurs late in the course of treatment with antipsychotics (usually after 3 months or more), (2) may emerge initially after antipsychotic discontinuation or dosage reduction, (3) can often be reduced in severity by increasing the antipsychotic dosage, and (4) may persist from months to years, even in the absence of drug—all of which are hallmarks of tardive syndromes in general.

• ∙ Pseudoakathisia:Reference Munetz and Cornes^25, Reference Stubbs and Halstead²⁶ This was a term invented to describe a condition in which there are the objective signs of akathisia without the subjective component. Because akathisia occurs commonly in patients with psychiatric disorders such as schizophrenia, it is not clear if pseudoakathisia should actually be considered a “pseudo” form of the disorder, because many patients experience discomfort that they are unable to understand or express clearly. Thus, we, like others,Reference Sachdev²¹ are skeptical about the existence of pseudoakathisia as a true subtype of akathisia, as opposed to one where other factors may simply reduce or alter the subjective complaints.

• ∙ Bing–Sicard akathisia:Reference Bing²⁷ This is a term that has been used to describe the occurrence of akathisia in parkinsonian disorders such as Parkinson’s disease (PD) and post-encephalitic parkinsonism, discussed below.

Etiology and Epidemiology

Although most commonly considered a side effect of antipsychotic medications, akathisia was described long before the advent of these drugs, and can still occur in other conditions:

Risk Factors for Drug-Induced Akathisia

Although acute antipsychotic-induced akathisia is often thought of in conjunction with schizophrenia, it appears that patients with mood disorders, particularly bipolar disorder, may actually be at higher risk.Reference Brüne^60–Reference Kane, Barnes and Correll⁶² Other risk factors may include higher dose or rapid dose increase,Reference Miller, Hummer, Oberbauer, Kurzthaler, DeCol and Fleischhacker⁶³ traumatic brain injury,Reference Nordström, Michaëlsson, Gustafson and Nordström^64, Reference Stewart⁶⁵ cancer,Reference Kawanishi, Onishi and Kato⁴³ and possibly iron deficiency.Reference Brown, Glen and White^66, Reference Krieger and Schroeder⁶⁷ There is no clear evidence of risk associated with age, gender, or ethnic background.Reference Sachdev^56, Reference Chong, Mythily and Remington⁶⁸ Chronic or tardive akathisia may be associated with advanced age and female gender, just as is tardive dyskinesia.Reference Sachdev²¹

Recognition and Differential Diagnosis

Differential diagnosis

There are several conditions that are commonly confused with akathisia:

Anxiety

Anxiety can be extremely difficult to distinguish from akathisia. Subjectively, patients with anxiety are less likely to complain about a need to move, or about how movement reduces the severity of their discomfort. Objectively, anxiety is associated with a greater degree of autonomic arousal, and observable excesses in movement usually manifest in upper extremity and facial fidgetiness, which is less common in akathisia.

Agitation with medical conditions

Agitation can occur in a variety of medical conditions, such as hyperthyroidism, sedative (including alcohol) or opiate withdrawal, AIDS, respiratory alkalosis, pre-eclampsia and eclampsia, meningitis, septicemia, pulmonary edema, and hyponatremia. These diagnoses can usually be made by their individual clinical features, and the agitation that accompanies them does not have a preferential lower extremity involvement, feeling of a need to move, or relief with movement.

Psychomotor agitation

This type of agitation is considered to be part of the primary presentation of a mental disorder, such as depression, bipolar disorder, schizophrenia, or post-traumatic stress disorder (PTSD), but it is not clear if “psychomotor agitation” relates to the same underlying mechanism in these different disorders. Unlike anxiety, which is a symptom, agitation is a sign, and hence a clinically observable entity, and the two may or may not be associated. Akathisia is often described as a type of agitation, but can usually be differentiated by the specific movement-impelling aspect of the subjective discomfort and the preferential involvement of the lower extremities when mild or moderate in severity.

Tics and Tourette disorderReference Lohr and Wisniewski^16, Reference Madruga-Garrido and Mir⁷³

The tics of Tourette disorder are divided into motor and phonic types, with motor tics typically appearing as rapid movements of eyes, lips, shoulders, fingers, and other areas, and phonic tics typically involving coughs, grunts, whistles, and animal noises. Although tics can involve the lower extremities, they are much more common in the face and upper extremities. Tics are often complex in nature, consisting of connected strings of separate motor and phonic tics. Some tic disorders, particularly Tourette disorder, are associated with a sensory disturbance that precedes the actual tic, sometimes called “a premonition” or “an itch,” which builds in intensity until it is relieved when the tic occurs. Thus, Tourette disorder is also a sensorimotor disorder, like akathisia. The upper torso distribution and the presence of phonic tics are distinguishing features from akathisia. Many patients with Tourette disorder may develop akathisia because of treatment with dopamine blocking drugs, which can serve to worsen the tics.Reference Weiden and Bruun⁷⁴ Patients who have both tics and akathisia can readily tell the difference in the subjective experience, as tics are not experienced as restlessness, but rather as a buildup of an uncomfortable feeling in a specific body part that culminates in a motor or phonic tic. There is also a tardive form of Tourette disorder that follows the course of tardive dyskinesia and is usually seen accompanying the more typical signs of tardive dyskinesia.Reference Fountoulakis, Samara, Siapera and Iacovides^75, Reference Stahl⁷⁶

For the remaining discussion of the differential diagnosis of antipsychotic-induced movement disorders, we have prepared Figure 1, which illustrates the relative preferred somatic region involved in mild to moderate cases of antipsychotic-induced akathisia, tardive dyskinesia, dystonia, and parkinsonian tremor. This differential distribution is often extremely helpful in making a diagnosis.

Figure 1

Relative prevalence of body region involvement of different antipsychotic-induced EPS, when they are present with mild to moderate severity. It can be seen that the preferred regional distribution can be of use in differentiating akathisia from other antipsychotic-induced conditions.

Antipsychotic-induced tardive dyskinesia (TD)Reference Burkhard^77–Reference Kahn, Munetz, Davies and Schulz⁸²

Because akathisia frequently occurs secondary to treatment with antipsychotic drugs, it is important to distinguish it from other antipsychotic-induced movement disorders. Perhaps the most commonly confused disorder is TD, which occurs late in the course of treatment and with which tardive akathisia is often associated. TD can be differentiated from acute akathisia on the basis of the timing of the appearance (late vs early), the nature of the movements, as well as the preferred somatic distribution of the movements (see Figure 1). TD consists of writhing athetoid or choreo-athetoid movements predominantly of the lower face and distal extremities (although it can involve the trunk, and the pharynx and diaphragm when severe). When the legs are involved, the movements are usually more pronounced in the toes than the more proximal muscles.

Antipsychotic-induced dystoniaReference Burkhard^77–Reference Sachdev⁸¹

Acute antipsychotic-induced dystonias (also called dystonic reactions) occur very soon, within 24-48 hours, after initiating or increasing antipsychotic dosing. There is also a tardive form, which shares the timing characteristics with TD (described earlier in the tardive akathisia section). Dystonias tend to involve the eyes (oculogyric crises), the tongue, the neck (torticollis), the shoulders, and trunk (see Figure 1).

Antipsychotic-induced parkinsonian tremorReference Burkhard^77–Reference Sachdev⁸¹

Although when assessed carefully, parkinsonian tremor can involve virtually all body regions to some extent, it is usually most clinically obvious in the hands and wrists when at rest (see Figure 1), where it assumes a pill-rolling characteristic consisting of alternating contractions of agonist and antagonist muscles in a very rhythmic rate of 4–7 cycles/second.

Akathisia vs Restless Legs Syndrome (RLS)

We are considering RLS separately from the other differential diagnostic entities because it is not clear whether akathisia and RLS truly represent different disorders. To see how difficult it can be to distinguish from akathisia, let us examine Ekbom’s original descriptionReference Ekbom⁸³:

RLS has more alternative names than most other medical disorders, being also known as Ekbom syndrome, Wittmaack–Ekbom syndrome, Willis–Ekbom disease, anxietas tibiarum, and anxietas tibialis. Ekbom himself originally called the disorder asthenia crurum paresthetica, highlighting the uncomfortable paresthesias in the lower extremities. RLS appears to be quite common and may affect 3–9% of the population, being more common in women, with advanced age, and with iron deficiency.Reference Trenkwalder, Paulus and Walters⁸⁴ The critical features of RLS are (1) a desire to move the extremities usually associated with some definable discomfort, (2) motor restlessness, (3) worsening of symptoms at rest with at least temporary relief by activity, and (4) worsening of symptoms later in the day or at night. As can be seen, apart from criterion 4, there is essentially no definitional difference between akathisia and RLS. This association with night-time and sleep is the essential clinical difference between RLS and akathisia, and was noted by Ekbom:Reference Ekbom⁸³

The relationship between akathisia and RLS often receives very little attention in discussions of the differential diagnosis of either of these conditions, which is surprising considering the large degree of clinical overlap between them. Part of the reason for this may be related to the different heritage of these concepts, with RLS studies coming more from the neurological and sleep literature, and akathisia from the psychiatric and psychopharmacological literature. An example of the diagnostic confusion is seen with PD, for example, where in some instances the restlessness is referred to as RLS,Reference Angelini, Negrotti, Marchesi, Bonavina and Calzetti^85–Reference Rijsman, Schoolderman, Rundervoort and Louter⁸⁹ and in others, akathisia,Reference Blaisdell^90, Reference Poewe and Hogl⁹¹ but it is unclear if there is a real distinction present. For example, it is very confusing diagnostically when investigators report the occurrence of akathisia at nighttimeReference Linazasoro, Massó and Suárez⁹² and RLS during the day.Reference Takahashi, Ikeda, Tomida, Hirata, Hattori and Inoue⁹³

It has been suggested that RLS can be distinguished from akathisia in that RLS occurs more commonly at night, is associated with greater sleep disturbance, is marked by myoclonus when severe, is exacerbated by lying down (akathisia may be somewhat relieved by lying down in some patients), and is experienced more as paresthesias (akathisia may be experienced more as restlessness).Reference Walters, Hening, Rubinstein and Chokroverty⁹⁴ However, none of these distinctions forms a firm basis for differentiating the conditions, and in fact RLS and akathisia may actually be closely related disorders, if not in fact variants of the same underlying pathophysiology. One distinguishing feature usually not addressed is that the two conditions may have different treatment responses. Although both have been described as responding to benzodiazepines, the use of dopamine agonists such as l-DOPA and opiates is standard treatments for RLS, but rarely if ever used for akathisia, whereas anticholinergic drugs and beta-adrenergic blockers are often used for akathisia (though the evidence is inconsistent; see below), but they are rarely used for RLS. These differences may not so much relate to differences in the underlying conditions, however, as to the fact that akathisia commonly occurs in individuals with psychiatric disorders, where the use of dopaminergic agonists may exacerbate manic or psychotic symptoms,Reference Marsden and Jenner⁹⁵ and the use of opiates is usually avoided because of abuse potential.

We should also mention periodic limb movement disorder (PLMD), which is also known as periodic leg movements of sleep (PLMS), which is a condition that clinically overlaps and is frequently associated with RLS. In fact, it has been considered to be simply a milder form of RLS (in which only the sleep component occurs) by some investigators. However, because RLS and PLMD can occur independently of one another, and there are differences in the sleep architecture between RLS and PLMD, they may actually represent different conditions.Reference Eisensehr, Ehrenberg and Noachtar⁹⁶ In any case, PLMD is different from akathisia in that it occurs during sleep, whereas akathisia is a condition of wakefulness.

Pathophysiology

The pathophysiology of akathisia is unknown. Because akathisia is so commonly observed in the context of treatment with first-generation antipsychotic medications and in parkinsonian states, it is reasonable to assume that dopamine blockade may be involved in its genesis, but this is not certain. More than 30 years ago, Marsden and JennerReference Marsden and Jenner⁹⁵ suggested that akathisia may be cause by a blockade of mesocortical dopaminergic pathways, which is still a viable but unproven hypothesis.

Stahl and LoonenReference Stahl and Loonen⁹⁷ have suggested another mechanism for akathisia, in which a generalized reduction in dopamine in the brain (such as is seen with antipsychotics and PD) may trigger compensatory mechanisms, particularly in the form of increased noradrenergic activity from the locus ceruleus. Because these adrenergic fibers innervate the shell portion of the nucleus accumbens to a greater extent than the core portion, a mismatch may occur resulting in dysphoric feelings and semipurposeful movements. This would also help explain the possible utility of beta-blocking drugs and the fact that akathisia is not seen in everyone when treated with first-generation antipsychotic drugs, which powerfully block dopamine receptors.

It is possible that mechanistic clues for akathisia can be found in studies of RLS. In the case of RLS, it has been suggested that dopamine reduction in the substantia nigra or in the dopamine A11 neurons (which originate in the hypothalamus and terminate in the spinal cord) may be involved. Recently, however, it has been observed that domperidone, which is a dopamine receptor antagonist that does not cross the blood–brain barrier, can cause a worsening of RLS in PD, suggesting that peripheral dopaminergic factors may be involved.Reference Rios Romenets, Dauvilliers and Cochen De Cock⁹⁸ Because of the previously mentioned similarity of RLS and akathisia, these possible mechanisms may be worth studying in akathisia as well.

Assessment

Currently, akathisia is a diagnosis made purely by clinical observation and patient report, as there is no confirmatory blood test, imaging assessment, or neurophysiological study available. The most commonly used tool for assessment is the Barnes Akathisia Rating Scale (BARS),Reference Barnes⁹⁹ which is a 4-item scale in which the subjective and objective components of the condition are rated separately, then combined. Each item is rated on a 4-point scale (0 to 3). There is 1 item that assesses objective signs, 2 subjective items that assess awareness and distress related to restlessness, and a global clinical assessment (see the Appendix).

When assessing the severity of akathisia, it can be quite helpful to observe patients when they do not know they are being evaluated, such as when sitting in the waiting area. This is because many individuals can exert a great deal of suppressive control on the motor aspects of the condition, especially when it is only of mild to moderate severity, causing a falsely low estimate of its severity.

Treatment

It is unfortunate that the actual evidence base for treatments of akathisia is very small, and many of the recommendations are based on clinician experience, case reports, and reports from studies of drugs where akathisia was not a primary concern. Nevertheless, the absence of strong evidence does not mean we should not try to address and treat the situation to the greatest possible extent. In our opinion, first and foremost, the cause of the akathisia should be identified and initial treatment aimed at that cause. Since the majority of patients with akathisia probably developed it secondary to psychotropic medications, the initial recommendation is, if possible, to reduce or switch medications. In patients on first-generation drugs, an attempt should be made to switch to second-generation agents that appear to cause less akathisia, including quetiapine and iloperidone. Also, if iron deficiency is present, then correcting this may be of some benefit. It should be remembered that occasionally, withdrawal akathisia can occur, so one should not judge the effectiveness of a dosage reduction or a medication switch for up to 6 weeks or more, as the akathisia may show a temporary exacerbation. If the akathisia worsens but does not disappear, then tardive akathisia may be present, and the general treatment guidelines follow those of acute akathisia below, with the exception that, when severe, tardive akathisia can be suppressed with a reinstitution of the offending agent at the previous dose.

After these maneuvers, if the akathisia is still present, it may be very difficult to treat. A host of different medications has been reported to be useful, but without much corroboration or evidence. These include anticholinergic drugs (such as biperiden, trihexyphenidyl, and benztropine),Reference Lima, Weiser, Bacaltchuk and Barnes^100–Reference Baskak, Atbasoglu, Ozguven, Saka and Gogus¹⁰² beta-blockers (such as propranolol and metoprolol),Reference Lima, Bacalcthuk, Barnes and Soares-Weiser¹⁰³ and serotonin 5-HT2A antagonists (such as mianserin, mirtazapine, and cyproheptadine).Reference Poyurovsky¹⁰⁴ There have also been reports of success with vitamin B6, n-acetylcysteine,Reference Berk, Copolov and Dean¹⁰⁵ and tetrabenazine,Reference Guay^106, Reference Jankovic and Clarence-Smith¹⁰⁷ although tetrabenazine is also reported to cause akathisia. Benzodiazepines have proven useful, but do not appear to relieve the akathisia as much as any associated anxiety.Reference Poyurovsky¹⁰⁴ Interestingly, two of the most commonly used treatments for akathisia, anticholinergic agents and beta-blockers, have both been reported to have poor evidence of efficacy in Cochrane reviews.Reference Lima, Weiser, Bacaltchuk and Barnes^100, Reference Rathbone and Soares-Weiser^101, Reference Lima, Bacalcthuk, Barnes and Soares-Weiser¹⁰³

By and large, after adjusting any offending agents (or treating any underlying conditions), there is little in the way of incontrovertible evidence for any therapeutic approach, and the treatment of akathisia becomes largely empirical, with many of the more severely affected patients ending up on double- and triple-drug regimens involving combinations of anticholinergic, beta-blocking, and serotonin blocking drugs, along with benzodiazepines. Occasionally, clonidine has been recommended, and there is a report of tardive akathisia responding to clonidine.Reference Nishikawa, Koga, Uchida and Tanaka²⁴ Rarely, dopamine agonists such as bromocriptine and amantadine have been tried.Reference Blaisdell⁹⁰ Finally, piracetam,Reference Fehr, Dahmen, Klawe, Eicke and Szegedi¹⁰⁸ buspirone,Reference Blaisdell⁹⁰ and opiatesReference Blaisdell⁹⁰ have been suggested to be of benefit in case reports of more serious akathisia.

Summary

Akathisia is a complex, confusing, and under-recognized problem that causes considerable suffering. It is extremely important for the clinician to be alert to its presence, as addressing akathisia in terms of alterations of current medication regimens can do much to relieve discomfort and to improve patient adherence, thereby reducing psychiatric and medical morbidity.

Disclosures

The authors do not have anything to disclose.