Akathisia and Second-generation Antipsychotic Drugs

Akathisia and Second-generation Antipsychotic Drugs

Rajeev Kumar; Perminder S. Sachdev

Authors and Disclosures

Published: 06/02/2009

Abstract and Introduction

Abstract

Purpose of Review: Akathisa is one of the most common and distressing neuroleptic-induced extrapyramidal side effects. Although it is well recognized in the context of conventional antipsychotic medications, there have been recent concerns raised by clinicians and researchers that this syndrome is overlooked in relation to second-generation or atypical antipsychotics. This review examines the recent literature relevant to second-generation antipsychotic (SGA)-induced akathisia.

Recent Findings: Recent studies using large databases clearly indicate that extrapyramidal side effects, in particular akathisia, do occur with the SGAs, although the frequency is not as high as with the conventional antipsychotics. Risk factors include use of high doses, high potency SGAs, or combinations of SGAs with other psychotropic drugs, bipolar depression, palliative care settings, and comorbid substance abuse in psychosis. The dopamine hypothesis remains plausible for understanding the pathophysiology of akathisia. There is emerging evidence that mirtazapine may be useful in the treatment of acute akathisia.

Summary: Even though akathisia is less prevalent with SGAs than with the first-generation drugs, it remains clinically important and all clinicians should be conversant with its recognition and management.

Introduction

Second-generation antipsychotics (SGAs), now the mainstay of antipsychotic treatment in most countries, have the shared feature that they produce fewer extrapyramidal side effects (EPSEs) than the conventional or first-generation antipsychotics (FGAs).[1] In fact, this relative lack of EPSEs is considered to be the defining feature of their 'atypicality'.[2] However, SGAs are not free of EPSEs and the whole range of EPSEs generally associated with FGAs has also been associated with the SGAs. In order to better understand the differential rates of EPSE production by the SGAs, a new classification has been suggested based on the dopamine D2 receptor binding affinity concept.[3] In this classification, clozapine and quetiapine are included in the low affinity group, olanzapine in the middle affinity group, and risperidone, ziprasidone, and aripiprazole in the high affinity group. Furthermore, it is suggested that olanzapine and risperidone at higher doses have high D2 occupancy, whereas quetiapine and clozapine do not.

We agree with a recent editorial[4] that the widespread and appropriate attention to metabolic effects with SGAs should not be a reason to ignore the problem of EPSEs, even though they may be mild or less frequent. In this update, we focus on akathisia (from Greek, literally 'not to sit'), one of the most common and disabling side effects of antipsychotics and some other drugs.[5,6] The various forms of akathisia have been well characterized in earlier studies, which include acute, chronic, withdrawal and tardive subtypes.[7,8] Because akathisia differs from Parkinsonian symptoms such as rigidity, bradykinesia, and tremor in its risk factors and treatment response, its pathogenetic mechanisms are also likely to be different, suggesting that drugs that do not produce significant EPSEs may yet cause akathisia. We examine the recent empirical evidence relating akathisia to SGAs and examine the risk factors, pathophysiology, and recent attempts at its treatment.

Previous meta-analyses of randomized double-blind controlled studies[9,10] found that atypical antipsychotics were less likely to produce EPSEs than FGA drugs. As a result of such informative accounts, several clinical guidelines have been developed and recommended that atypical antipsychotics should be used as first-line agents in the treatment of schizophrenia and related disorders.[11] One major criticism of such studies has been the use of high doses of haloperidol as a comparator drug, thus overstating an advantage for atypical over conventional antipsychotic drugs. Rates of akathisia are available from a number of treatment studies using SGA drugs, both in comparison with placebo and with other SGAs and some FGAs

Clozapine

In an older study,[12] clozapine was compared with chlorpromazine in 151 hospitalized patients with schizophrenia. Twelve percent of clozapine patients developed EPSE-related adverse effects compared with 25% in the chlorpromazine group. A later study[13] used the Extrapyramidal Rating Scale (ERS) and extracted the akathisia items from it in 29 patients receiving FGAs and 23 patients receiving clozapine. All patients had a diagnosis of a psychotic disorder. Akathisia was reported to be present in 39% of clozapine-treated patients compared with 45% of patients treated with FGAs. A rare case of acute nocturnal akathisia induced by clozapine has been reported recently in a 43-year-old man with a diagnosis of schizophrenia.[14] The condition was successfully treated with propranolol 40 mg/day while the patient continued 350 mg/day of clozapine. So it is clear that, although clozapine does not increase the risk of Parkinsonism over placebo, akathisia still occurs at a higher frequency. A caveat in relation to clozapine studies is that patients have previously been treated with other antipsychotics and some may have persistent akathisia attributable to the earlier drugs.

Risperidone

In a Cochrane review, it was reported that about 33% of patients with schizophrenia developed EPSEs on risperidone or olanzapine.[15] However, 25% of patients receiving risperidone required medication to alleviate the akathisia, which was significantly higher than those receiving olanzapine during a 12-week to 12-month trial period. The average dose of risperidone ranged from 1.5 to 10 mg/day and from 5 to 30 mg/day for olanzapine. The generally recommended dose for risperidone (up to 6 mg/day) is much less than the upper range of 10 mg/day, with the drug losing its EPSE advantage at the higher doses. A similar rate of 27.5% EPSEs was reported in another review in comparison with 12.8% of placebo-treated patients with autism spectrum disorders.[16] A review based on published studies on the safety and efficacy of risperidone augmentation of clozapine in clozapine treatment-resistant patients showed that 9.3% of patients reported EPSEs or akathisia.[17] The low rate of akathisia could be attributed to the use of low-dose risperidone.

Olanzapine

A pooled analysis of four randomized open-label studies in the Middle East and North Africa on olanzapine versus chlorpromazine in the treatment of schizophrenia was reported recently.[18] The dose ranged from 5 to 20 mg/day for olanzapine (n = 83) and from 200 to 800 mg/day for chlorpromazine (n = 40). Akathisia was reported in 2.4% of the olanzapine group compared with 10% in the chlorpromazine group. The European Mania in Bipolar Longitudinal Evaluation of Medication (EMBLEM) observational study[19] reported the effectiveness and tolerability of olanzapine monotherapy and olanzapine combination therapy with other antipsychotics, anticonvulsants, and/or lithium in the treatment of mania after a 12-week trial. The incidence of akathisia was 3% in the olanzapine monotherapy group compared with 6% in the olanzapine combination group, a statistically significant difference.

Quetiapine

Prevalence of EPSEs and akathisia in quetiapine-treated bipolar manic patients has been reported based on four randomized placebo-controlled double-blind trials.[20] Studies evaluated quetiapine monotherapy (800 mg/day), versus placebo, with lithium or haloperidol monotherapy as controls, quetiapine in combination with a mood stabilizer (lithium or divalproex) compared with placebo and a mood stabilizer. Studies used the Simpson-Angus Scale (SAS) and Barnes Akathisia Scale (BAS) for assessment of EPSEs and akathisia, respectively. There was no difference in the incidence of EPSEs between quetiapine monotherapy and placebo (12.9 versus 13.1%), quetiapine in combination with lithium/divalproex and placebo in combination with lithium/divalproex (21.4 versus 19.2%). The incidence of akathisia was 3.3% with quetiapine monotherapy compared with 6.1% with placebo. Even the combination of quetiapine with lithium/valproex showed an incidence of only 3.6% compared with 4.9% with placebo and lithium/divalproex combination. The study concluded that quetiapine treatment was similar to placebo in terms of the occurrence of akathisia in bipolar patients. In a study of switching from previous antipsychotics to quetiapine in patients with schizophrenia and preexisting EPSEs,[21] the authors noted a significant reduction in Parkinsonism and akathisia

Iloperidone and Ziprasidone

A large 4-week, randomized, placebo-controlled, multicentre study comparing the efficacy and safety of iloperidone with ziprasidone and placebo[22] showed that iloperidone was associated with lower incidence of EPSEs, particularly akathisia compared with ziprasidone. These findings suggest that drugs such as iloperidone with mixed D2/5-HT-2 antagonism may be superior to other SGAs in terms of their EPSE profile. Sudden emergence of akathisia following ziprasidone dose reduction has been reported in four female patients with bipolar disorder, alerting the clinicians that SGAs may be associated with sudden EPSEs while they are taking or discontinuing their drugs.[23] Although withdrawal akathisia has been reported with FGAs, controlled studies are lacking with SGAs.

Aripiprazole

The efficacy and safety of aripiprazole as an adjunct therapy in major depressive disorder showed that akathisia occurred in 4.5% of those patients who took an antidepressant and an adjunct placebo, compared with 23.1% of those who took an antidepressant and adjunct aripiprazole,[24] although only one patient discontinued due to akathisia. A Cochrane systematic review of aripiprazole versus FGAs showed that aripiprazole was superior to typical antipsychotics in terms of occurrences of EPSEs and tolerability.[25] The relative risk (RR) of akathisia in 897 patients from three randomized controlled trials was calculated as RR 0.39 [confidence interval (CI) 0.3-0.6], number needed to treat (NNT) 11 (CI 14-9). In an open-label, rater-blinded, aripiprazole augmentation study on treatment-resistant depression,[26] akathisia was reported in 20% of patients, suggesting that antidepressants in combination with FGAs may increase the incidence of akathisia. Another study using aripirazole in refractory bipolar depression[27] found that 42% of patients had treatment-related akathisia, again suggesting that SGAs in combination with other psychotropic medications will increase the risk of development of akathisia.

Amisulpride

The effectiveness and tolerability of amisulpride was compared with that of risperidone for the treatment of behavioural and psychological symptoms in patients with Alzheimer's disease.[28] EPSEs were reported to be 20% in the amisulpride group compared with 38.4% in the risperidone treatment group. The average dose for amisulpride was 85 mg/day and for risperidone 1.8 mg. There was no statistically significant difference between amisulpride and risperidone in terms of the occurrence of akathisia as measured by BAS (0.3 ± 0.5 versus 0.2 ± 0.4). It is worth noting that the study used very low doses of risperidone and amisulpride.

Naturalistic and Large Studies Using Several Second-generation Antipsychotics and Low Potency First-generation Antipsychotics

FGAs were compared with SGAs in a naturalistic setting in Finland.[29] The authors studied 100 patients with psychosis who were attending a polyclinic. Akathisia was rated using BAS and antipsychotic doses were converted to chlorpromazine equivalents. A total of 17 patients (17%) were detected to have akathisia. Of particular importance is that none of the patients in the SGA groups had akathisia. In the FGA groups, the prevalence of akathisia was marked with the highest rate of 44% in the depot FGAs and selective serotonin reuptake inhibitor (SSRI) group. Unfortunately, the names of individual FGAs or SGAs were not provided in this study. Additionally, the conversion of drug doses to chlorpromazine equivalents can be questioned for its accuracy.

The Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study is a major advance in the study of efficacy and tolerability of antipsychotics in a real-life clinical setting. An earlier report on this study showed no difference in the prevalence of drug-induced Parkinsonism, akathisia, or any abnormal movements in patients receiving olanzapine, quetiapine, risperidone, ziprasidone, or perphenazine.[30] However, it did show that akathisia still was a problem and occurred at a frequency of 5% in patients treated with olanzapine or quetiapine, 7% for risperidone or perphenazine, and 9% for ziprasidone. The comparable rates of akathisia with perphenazine and SGAs could be related to a sampling bias in the CATIE study, in which patients with tardive dyskinesia were not included in the perphenazine group. A subsequent more rigorous analysis of the CATIE data also showed that the incidence of treatment emergent Parkinsonism, akathisia, and tardive dyskinesia was not significantly different between moderate dosages of perphenazine (an intermediate potency conventional antipsychotic) and the four newer forms of atypical antipsychotics.[31] However, it is worth noting that there was a trend towards more patients on risperidone and perphenazine receiving medications for treatment of akathisia. In a recent double-blind comparison study of FGAs and SGAs (Treatment of Early Onset Schizophrenia Spectrum Disorders - TEOS study),[32] researchers compared the efficacy and safety of olanzapine (2.5-20 mg/day), risperidone (0.5-6 mg/day), or molindone (10-140 mg/day), and 1 mg/day of benztropine for 8 weeks. The adverse effects profile of these drugs showed that 18% of patients on molindone, 13% on olanzapine, and around 8% on risperidone had akathisia rated using the BAS.

The differential effects of EPSEs with SGAs were studied in patients with bipolar disorders.[33] The authors classified the SGAs into low potency (quetiapine, olanzapine) and high potency (risperidone, ziprasidone, aripiprazole) groups. They assessed 51 individual patient trials for a mean duration of 25.5 weeks. EPSEs were assessed using the Abnormal Involuntary Movements Scale (AIMS), BAS, and SAS. Results showed that 62.7% of trials resulted in moderate-to-severe EPSEs. There were no differences between high and low-potency agents in terms of EPSE frequencies or discontinuation rates. However, in a multiple regression model, akathisia was found to be less common in low-potency drugs. In addition, patients in the younger age group were more likely to have akathisia. This important study highlights the fact that EPSEs occur at a higher frequency than reported by clinical trials.

Longitudinal studies focusing on akathisia, both for its longitudinal course and delayed or tardive onset with SGAs, are generally lacking. In a recent longitudinal study, Modestin et al.[34] reported an interesting trend of akathisia occurring at a rate of 14% during both their baseline assessment in 1995 and follow-up in 2003/04. However, they noted that patients who took clozapine improved remarkably, whereas the condition of patients on other atypicals (olanzapine, risperidone, quetiapine, and sertindole) became worse. Although there are methodological issues in interpreting the results, the study underscores the fact that atypical antipsychotics continue to produce significant akathisia during long-term use, with the likelihood that although some patients improve, others develop akathisia with a delayed onset. A number of methodological issues need to be considered in studies that assess akathisia. First, comparing EPSE rates from different trials cannot adequately account for population (e.g. age, sex, diagnosis, ethnicity, etc.) and methodological differences. Pooling data from multiple studies can further complicate this issue. Second, most studies are designed to have adequate power to detect clinical effectiveness rather than side effects, thus creating underpowered studies for the evaluation of side effects. Third, akathisia can often be mistaken for agitation related to the underlying psychiatric disorder, thereby creating an underreporting bias. It is known that akathisia can be intermittent or have a delayed onset, thereby missing the diagnosis during office visits or in short duration trials. Finally, randomized controlled trials do not usually reflect real-life clinical settings. Patients enrolled in the studies are highly selected and patients with various risk factors that can potentially increase the risk of EPSEs are often excluded.

Risk Factors

We systematically looked at the recently published studies related to risk factors for SGA-induced akathisia. In a systematic review conducted by Gao et al.[35] of studies on the frequency of occurrence of EPSEs induced by antipsychotics in patients with bipolar disorder and schizophrenia, it was concluded that bipolar-depressed patients are at increased risk of acute antipsychotic-induced movement disorders. The authors used number needed to harm (NNTH) to estimate the risk. In regard to akathisia, the NNTH was 7 for haloperidol in patients with schizophrenia compared with 17 for aripiprazole, 19 for olanzapine, 17 for quetiapine, and 35 for ziprasidone. In mania, the NNTH in relation to akathisia was 4 for haloperidol and 9 for aripiprazole, both suggesting an increased risk. However, the NNTH for quetiapine and ziprasidone was much higher. Of particular note was that a NNTH of 5 was observed for aripiprazole in patients with bipolar depression. These findings suggest that patients with bipolar mood disorders, in particular bipolar depression, are at increased risk of developing akathisia with both conventional and atypical antipsychotics. These findings are relevant because several atypical antipsychotics are currently used for the treatment of bipolar disorders.

Patients in palliative care units may develop several neuropsychiatric syndromes that include delirium, agitation, and other behavioural disturbances due to various types of neurological and neuropsychiatric conditions. In addition, they also develop nausea and other general medical disturbances. Clinicians use a number of medications in the palliative care setting, which include antiemetics (metoclopramide, prochlorperazine), antipsychotics, antidepressants (tricyclics, SSRIs) and calcium channel blockers (cinnarizine, flunarizine, diltiazem), all of which can cause significant akathisia.[36] If the akathisic symptoms are misdiagnosed in these settings, patients might be further treated for agitation and subsequently be prescribed antipsychotics, which would further aggravate akathisia. An association between akathisia and acculturation has been reported in an Australian study,[37] in which the authors found a 60% prevalence of akathisia in a community sample of patients treated with olanzapine, risperidone, fluphenazine, zuclopenthixol, trifluperazine, and haloperidol. The association between smoking and akathisia was reported in a recent study.[38] The authors explored the self-medication hypothesis, whereby smoking would reduce akathisia in a sample of 250 outpatients with schizophrenia. Using the BAS, the authors noted that heavy smoking was not associated with akathisia (41% of patients with akathisia were heavy smokers versus 39% of patients without akathisia), arguing against the hypothesis. Psychoactive substances may interact with antipsychotics causing EPSEs or worsening of EPSEs. The results so far have been inconsistent and this may be partly due to methodological limitations. In a recent study, Potvin et al.[39] addressed this issue by investigating the effects of psychoactive substances on EPSEs in a sample of 41 patients with dual diagnosis schizophrenia. All patients were treated either with clozapine or quetiapine for at least 4 weeks. EPSE instruments used were the Extrapyramidal Symptoms Rating Scale (ESRS) and the BAS. In this well controlled study, patients with dual diagnosis were more likely to show higher scores on subjective EPSE complaints and higher ESRS scores than the schizophrenia only group. Of note, patients with a dual diagnosis had more Parkinsonian signs, and a subgroup analysis showed that patients abusing cocaine had more EPSE complaints, EPSEs, Parkinsonism, and signs of akathisia. As this was a cross-sectional study, one cannot determine the directionality of the cause and effect

The Subjective Experience of Akathisia

Some recent studies have looked at the subjective experience of akathisia. Sixty-seven outpatients with schizophrenia treated with risperidone or haloperidol were assessed for akathisia (BAS) and their subjective cognitive dysfunction (Frankfurt Compliant questionnaire).[40] The results showed that akathisia was significantly correlated with a number of subjective cognitive-perceptual deficits. These included anxiety, disorder of selective attention, perceptual disorder, and disorder of coping responses. The authors highlighted the importance of early therapeutic interventions for akathisia to reduce subjective cognitive dysfunction and impairment of coping. The subjective experience of akathisia was further highlighted in a series of four cases reported recently.[41] The author emphasized that side effects such as akathisia can be misinterpreted by the patients and adequate attention should be paid to the patients' experiences and supportive as well as cognitive behaviour therapy should be used to reduce such experiences. It has also been noted that akathisia occurring early in treatment or after increases in doses may be more troublesome and distressing for the patents.[42] Chronic akathisia from risperidone has been reported recently as a reason for a patient to become extremely irritable, easily agitated, and sometimes violent.[43]

Second-generation Antipsychotic-induced Akathisia in Children and Adolescents

In a large naturalistic sample of adolescent patients with schizophrenia treated predominantly with SGAs, Gebhardt et al.[44] reported that akathisia assessed by BAS significantly correlated with the subscale items of hostility and suspiciousness suggesting that movement disorders and psychopathology may have shared anatomical/pathophysiological mechanisms. However, this may be related to problems in the diagnosis of akathisia and distinguishing it from other symptoms. In an open-label, randomized comparison of olanzapine versus risperidone in the treatment of childhood onset schizophrenia, 25 children were assessed for efficacy and tolerability.[45] The average dose range for risperidone was 0.25-4.5 mg/day and 2.5-20 mg/day for olanzapine. Akathisia assessed by the BAS was not significantly different in the two groups, suggesting that the two drugs did not differ in terms of their EPSEs in children with schizophrenia.

Pathophysiology of Akathisia

Although there are many possible hypotheses for the pathophysiology of acute akathisia, none is completely satisfactory. So far the most attractive hypothesis is dopamine receptor blockade in the mesocortical and mesolimbic regions of the brain. It is unlikely that a single neurotransmitter hypothesis will explain all the complex features of the disorder, and the interaction of several neurotransmitters may be involved. There have been some recent imaging, genetic, and neurotransmitter depletion studies that have looked at the pathophysiology of akathisia. Striatal dopamine-2 (D2) receptor occupancy by antipsychotics has been implicated in the pathophysiology of EPSEs. In this regard, a recent study using single photon emission computed tomography (SPECT) and ligand iodobenzamide showed that bipolar patients receiving 5-45 mg/day of olanzapine for 2 weeks did not show any EPSEs at a D2 occupancy level of 28-80%, suggesting that, at clinically relevant doses, it is unlikely that bipolar patients would develop EPSEs.[46] In a recent study,[47] dopamine depletion using the administration of alphamethyl paratyrosine (AMPT) resulted in subjective changes in a group of patients with schizophrenia, including dysphoria, social withdrawal, and personal distress followed by akathisia, akinesia, and rigidity.

In a single case study,[48] olanzapine-induced akathisia was studied using 18F-fluoro-deoxyglucose-positron emission tomography (FDG-PET) during akathisia and after recovery. Results showed that akathisia was associated with reduced metabolic activity in the thalamus and cerebellum. More importantly, the metabolic activity recovered when akathisia disappeared after discontinuation of olanzapine. The role of the serotonin system was also explored in a study on the role of serotonin transporter promoter genotypes in acute antipsychotic efficacy and side effects in schizophrenia.[49] The authors used the SAS, BAS, and abnormal involuntary movement scale for assessment of EPSEs. There was no significant association between EPSEs and the serotonin transporter promoter gene polymorphism.

Treatment

Although there has not been any major advance in the treatment of SGA-induced akathisia, all akathisia treatment trials are relevant, as the findings may be applicable to SGA-induced akathisia. So far the established treatments for neuroleptic-induced akathisia include anticholinergic and antiadrenergic drugs in addition to a dose reduction strategy if possible. A recent Cochrane review on the evidence of anticholinergics in neuroleptic-induced akathisia concluded that there was insufficient data based on good quality research to support the use of anticholinergics.[50] However, we suggest that anticholinergics should not be abandoned in the treatment of akathisia.[6] One randomized trial of multiple doses of diphenhydramine as a prophylactic agent in metoclopramide-induced akathisia in an emergency setting has been reported.[51] The results of the trial indicated that there was no advantage of routine prophylactic use of diphenhydramine, but for patients who took 20 mg of metoclopramide, less subjective restlessness was reported. The role of mirtazapine in the treatment of akathisia was reviewed recently based on a limited number of publications.[52] So far there have been three case reports, one placebo-controlled trial, and one placebo and propranolol-controlled study. Pooled results showed that mirtazapine demonstrated an overall response rate of 53.8% compared with a 7.7% response rate for placebo and a 30% response rate for propranolol. One of the above studies was a randomized, double-blind, placebo-controlled and propranolol-controlled study, which needs some attention.[53] Ninety patients with antipsychotic-induced akathisia were randomly assigned to mirtazapine, (n = 30, dose 15 mg/day), propranolol (n = 30, dose 80 mg/day), or placebo (n = 30). A reduction of two points on the BAS was used as the primary outcome measure. Twenty-six percent of patients dropped out due to lack of response or adverse events. A reduction in the BAS score was found in 34% of patients in the mirtazapine group and in 29% of those in the propranolol group compared with placebo, which showed a reduction in 11%. Forty-three percent of patients in the mirtazapine-treated group and 30% in the propranolol-treated group responded to treatment compared with 6.7% in the placebo group. Five patents from the propranolol group dropped out due to hypotension or bradycardia compared with none in the mirtazapine group. The authors concluded that mirtazapine should be considered as a first-line treatment option for acute antipsychotic-induced akathisia, especially when propranolol is contraindicated in some patients. However, it is important to keep in mind that antidepressant medications, including mirtazapine, have been implicated in the genesis of akathisia. Another interesting trial compared the efficacy of intramuscular biperiden with isotonic saline in a double-blind randomized design[54] and noted that intramuscular biperiden had no advantage over isotonic saline in treating acute neuroleptic-induced akathisia. On the basis of the above trials, no firm conclusions can be drawn. Perhaps, the best approach now is to follow previously established clinical practice in the management of acute akathisia, which includes dose reduction and the use of drugs such as anticholinergics, antiadrenergic drugs, mianserin or ritanserin, and possibly mirtazapine. A case of aripiprazole-induced tardive akathisia responding to the dopamine agonist ropinirole has been reported recently.[

Conclusion

There is clear evidence that akathisia is a side effect of SGAs, though less frequently than with the conventional antipsychotics. The presence of akathisia can lead to noncompliance, personal distress, and increased suicide risk for the patients. Along with management of other side effects of SGAs, such as the metabolic syndrome, adequate attention should be paid to EPSEs, in particular akathisia.