Clinical predictors of aripiprazole-associated psychotic symptoms
Takeuchi et al14 aimed to establish predictors of worsening psychosis in a naturalistic setting where patients slowly transitioned to aripiprazole from previous antipsychotic treatment. Patients were required to be on a stable dose of an antipsychotic before participating in the study. Aripiprazole was started at 12 mg/d for 2 weeks with flexible dosing from weeks 2 to 52. Previous antipsychotic therapy was reduced biweekly by 25%. The incidence of worsening psychopathology after aripiprazole initiation was higher in the group of patients who had previously received high-dose antipsychotic therapy (average chlorpromazine equivalents [CPZE]: 727 mg/d) compared with the group on low dosages (average CPZE: 382 mg/d). It is possible that previous high-dose antipsychotic therapy was indicative of more significant baseline psychopathology; however, the worsened group and stabilized group had similar baseline Clinical Global Impressions-Severity scores.
Pae et al25 aimed to find predictors of worsening psychosis with aripiprazole in patients whose previous antipsychotic therapy was immediately discontinued. They found lower baseline disease severity was associated with significant worsening during the first month of aripiprazole treatment.
Other potential explanations
Aripiprazole’s manufacturer reported the incidence of psychosis-related adverse events in an analysis of 9 randomized schizophrenia trials.27 The rates of psychosis-related adverse events ranged from 0.6% to 18%, but there was no apparent relationship to study design or method of transitioning to aripiprazole. Rates of psychosis-related adverse events were similar between aripiprazole and the control group (placebo in 3 studies, another antipsychotic in 2 studies).
Emergence or worsening of psychotic symptoms temporally associated with aripiprazole initiation does not necessarily imply causation. As in Mr. N’s case, it is not always possible to determine whether worsening psychosis is the natural disease course or a treatment effect. In addition, it is not possible to differentiate lack of efficacy from a true propensity for aripiprazole to worsen psychosis.
It also is conceivable discontinuation or dosage reduction of a previous antipsychotic would worsen psychotic symptoms or cause side effects. When significant changes in psychopathology or side effects develop during the transition from 1 antipsychotic to another, it is difficult to determine etiology. Specifically, rapid transition from a medication with significant anticholinergic and antihistaminic properties—such as quetiapine or olanzapine—to 1 without these properties—such as aripiprazole—may result in symptoms of activation, including restlessness, insomnia, and anxiety. Consequently, these symptoms could be mistaken for worsening psychosis.28 Only 1 patient in this series was reported to abruptly discontinue an antipsychotic with significant anticholinergic properties (clozapine) before initiating aripiprazole.24 Studies by Takeuchi et al14 and Pae et al25 did not report the relative baseline use of antipsychotic medication with anticholinergic properties.
In a pooled analysis of treatment-emergent adverse events in 5 randomized clinical trials of patients receiving aripiprazole for acute relapse of schizophrenia, the incidence of akathisia was 10%, although it is not clear if this is a dose-related adverse effect.29 Because akathisia may be confused for worsening psychosis,30 it is possible akathisia was mistakenly identified as worsening psychotic symptoms in Mr. N’s case, as well as several reports from our literature review.
Covert akathisia is unlikely to explain worsening psychopathology observed in all patients in our literature review because confusion of akathisia and worsening psychosis is not a widespread phenomenon. In a post hoc analysis of pooled safety data from aripiprazole trials, Kane et al31 did not find a correlation between presence of akathisia and aripiprazole efficacy as measured by the Positive and Negative Syndrome Scale (PANSS) total, PANSS positive, PANSS negative, Clinical Global Impressions-Severity, Clinical Global Impressions-Improvement, and percentage of responders. Pae et al25 also noted there was no correlation between scores on the Barnes Akathisia Rating Scale and worsening psychopathology in patients switched to aripiprazole.
An antagonist always is an antagonist and clinicians have appreciated this concept since the days of chlorpromazine. The activity of aripiprazole, however, is on a pharmacologic continuum between a neutral antagonist and full agonist and currently there is no way to precisely determine the level of D2 receptor agonist action in a patient.
Although it is interesting to speculate that aripiprazole’s D2 receptor agonist action may contribute to worsening psychosis,32-34 there are other plausible explanations to consider. Rapid transition from a drug with significant anticholinergic properties and aripiprazole-associated akathisia may contribute to worsening psychopathology in patients starting aripiprazole. Because covert side effects may be incorrectly identified as psychotic agitation, we cannot exclude this as a possible etiologic factor in Mr. N’s case as well as the cases in our literature review.