Introduction

As a partial dopamine agonist, aripiprazole is pharmacologically unique amongst existing antipsychotic drugs (Burris et al. 2002). It is postulated that this bestows benefits from the standpoint of side effects related to dopamine blockade, including extrapyramidal symptoms (EPS) and hyperprolactinemia in patients with schizophrenia (Marder et al. 2003). Aripiprazole also has a favorable profile in terms of weight gain and associated metabolic disturbances (e.g., disturbances in lipids and glucose) (Marder et al. 2003).

In view of its unique pharmacological and clinical profile, randomized controlled trials have been conducted to investigate the impact of adding, or switching to, aripiprazole to address problems such as hyperprolactinemia (Shim et al. 2007) or metabolic side effects in patients with schizophrenia (Fleischhacker et al. 2010; Newcomer et al. 2008; Stroup et al. 2011). While these studies have reported benefits with such a strategy, there have also been reports of worsening of psychotic symptoms in patients with schizophrenia, for example after adding aripiprazole compared to placebo (Fleischhacker et al. 2010), or switching to aripiprazole versus (L69) continuing olanzapine (Newcomer et al. 2008). The possibility that aripiprazole may worsen psychotic symptoms in patients with schizophrenia dovetails with a body of research suggesting such an association for agents that increase dopaminergic activity (Jaskiw and Popli 2004; Lieberman et al. 1987).

On the other hand, these clinical trials did not provide detailed clinical information on individual patients who experienced a worsening of psychotic symptoms, and thus cannot allow for conclusions regarding aripiprazole and a causal relationship. To date, there have been a number of case reports of psychotic symptoms worsened by aripiprazole in patients with schizophrenia since the first such case was reported in 2004 (Ramaswamy et al. 2004). Contrary to the clinical trials, these cases supply enough information to not only evaluate the quality of the causal relationship between aripiprazole and worsening of psychotic symptoms but also identify the potential risk factors of this phenomenon. Based on these objectives, we carried out a systematic review of reported cases.

Method

A literature search was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) statement (Moher et al. 2009) on August 18, 2012, in the PubMed and the EMBASE using the following search terms: “aripiprazole AND schizophreni*”. Our literature search was limited to “case reports,” “humans,” and “English” in the PubMed, and “letters,” “human,” and “English language” in the EMBASE. The reference lists of the relevant reports were also examined.

Inclusion criteria were as follows: (1) diagnosis of schizophrenia or schizoaffective disorder; (2) worsening of psychotic symptoms in conjunction with aripiprazole, including hallucinations, delusions, disorganization, and/or bizarre behavior; and (3) aripiprazole ≤30 mg/day (i.e., maximum recommended therapeutic dose).

Each of the cases meeting inclusion criteria were screened for the following information: age and gender, diagnosis, onset or history of illness/treatment, rationale for initiating aripiprazole, previous antipsychotic medication(s)/dose(s), concomitant psychotropic medication(s)/dose(s), switching strategies to aripiprazole until psychotic worsening commences, duration until psychotic worsening commences since initiation of aripiprazole, nature of psychotic/non-psychotic symptoms that increased, effective/ineffective strategies for managing increased psychotic symptoms, and duration until psychotic worsening resolves.

Psychotic symptoms were defined by the following symptoms: hallucinations, delusions, disorganization, and bizarre behavior. Increased anxiety, hostility, and psychomotor activity were also recorded under the following categories, respectively: agitation, aggression, and activation.

The quality of the causal relationship between aripiprazole and increased psychotic symptoms was systematically evaluated for each case according to “the guidelines for evaluation of drug-associated events” developed by Aubry et al. (2000). These guidelines, developed to evaluate the relationship between atypical antipsychotics and mania/hypomania, were modified to specifically evaluate psychotic symptoms and are detailed in Table 1.

Table 1 The modified guidelines for evaluation of drug-associated events for psychotic symptoms worsened by aripiprazole in schizophrenia or schizoaffective disorder
Full size table

Items 1–7 of the original guidelines were maintained, with item 8 excluded based on the original definition (Aubry et al. 2000). In line with the original scale, the total number of fulfilled items was used to categorize each case as follows: 1–3 as “questionable,” 4 as “moderately suggestive,” and 5–7 as “highly suggestive” (Aubry et al. 2000).

Results

A total of 22 cases in 16 reports (Adan-Manes and Garcia-Parajua 2009; Ahuja and Lloyd 2007; Avari et al. 2011; Barnas et al. 2005; Burke and Lincoln 2006; Chiu et al. 2011; DeQuardo 2004; Glick et al. 2006; Grover et al. 2006; Kapusta et al. 2007; Lea et al. 2007; Letmaier et al. 2012; Ponde and Novaes 2007; Raja 2007; Ramaswamy et al. 2004; Reeves and Mack 2004) were identified (Fig. 1); a summary is provided in Table 2. Case reports have been published almost yearly since 2004 when aripiprazole became available in the USA.

Fig. 1
figure 1

PRISMA flow diagram of the literature search

Full size image
Table 2 Cases of psychotic symptoms worsened by aripiprazole in schizophrenia or schizoaffective disorder
Full size table

Case characteristics

Age ranged from 23 to 72 years (median, 48), and cases were evenly split based on gender. Most cases (15 cases) were diagnosed with schizophrenia, with the remainder (seven cases) schizoaffective disorder. Patients were characterized as being chronic and/or ill >10 years in most cases (15 cases).

The reasons for introducing aripiprazole included persistent symptoms (12 cases), side effects (seven cases), both (one case), or other reasons (two cases). Previous antipsychotic treatment included atypical antipsychotic monotherapy (15 cases), typical antipsychotic monotherapy (four cases), combinations of atypical and typical antipsychotics (two cases), and combinations of atypical antipsychotics (one case). The prior antipsychotic dose exceeded the recommended maximum therapeutic dose (Gardner et al. 2010) in seven cases, and did so by one daily defined dose (DDD) (WHO Collaborating Centre for Drug Statistics Methodology, available at http://www.whocc.no/atc_ddd_index/) in most cases (15 cases). Mood stabilizers, used in eight cases of 12 cases, represented the most frequent concomitant medication class. Anticholinergic drugs were used in two cases.

Psychotic symptoms worsened after adding aripiprazole to current treatment, left unchanged, in eight cases. In most of these (six cases), psychotic symptoms worsened at a dose before/without aripiprazole titration. Doses of aripiprazole associated with a worsening of psychotic symptoms ranged from 5 to 30 mg/day (mode, 15 mg/day), which suggests no clear relationship between aripiprazole dose and worsening of psychotic symptoms. Duration of aripiprazole exposure until increased psychotic symptoms ranged from days to months (median and mode, 2 weeks). Specifically identified symptoms included hallucinations (11 cases), delusions (19 cases), disorganization (two cases), and/or bizarre behavior (two cases). Intensity of increased psychotic symptoms was described as “significantly worsened/increased” (case 2, 4, and 20), “outbreak” (case 15), “dramatically increased” (case 17 and 21), “severely exacerbated” (case 22), “intensely paranoid” (case 5), or “overtly paranoid” (case 6). Other symptoms noted were agitation (10 cases), aggression (11 cases), and/or activation (seven cases).

The worsening of psychotic symptoms was not improved by increasing or decreasing aripiprazole, while leaving other medications unchanged, in approximately half the cases (six cases out of 13 cases). Further to this point, increasing aripiprazole dose was implicated in further worsening in four cases. On the other hand, discontinuation of aripiprazole, without changing other medications, resulted in resolution in most instances where this was implemented (eight cases). Just as time to onset of psychotic worsening was quite variable, duration until improvement of psychotic symptoms after effective intervention ranged widely, from days to months (median, 8 days; mode, 1 week). Table 2 provides details of each case.

Besides the 22 cases, 4 cases were identified as rechallenge trials (Table 2). In all, psychotic symptoms worsened after simply adding aripiprazole across a wide range of doses (2.5–15 mg/day) to the current regimen and resolved shortly after discontinuation of aripiprazole.

Quality of causal relationship between aripiprazole and psychotic worsening for cases

Causal relationship between aripiprazole and worsening of psychotic symptoms is summarized as follows: 11 cases, “highly suggestive”; three cases, “moderately suggestive”; and eight cases, “questionable” (Table 2). All four rechallenge cases were classified as “highly suggestive.”

Discussion

Our main findings from this systematic review are as follows: (1) approximately half of reported cases demonstrated a robust causal relationship between aripiprazole and a worsening of psychotic symptoms, (2) approximately one third of cases reported psychotic symptoms worsened after simply adding aripiprazole to the current treatment regimen, and (3) approximately one third of cases showed resolution of psychotic worsening was achieved by discontinuing aripiprazole without changing other medications. Aripiprazole also exacerbated psychotic symptoms across a wide range of doses. Taken together, the evidence seems to substantiate a causal relationship between aripiprazole and a worsening of psychotic symptoms in a proportion of patients with schizophrenia or schizoaffective disorder.

What is less clear is whether there are predictors of those at risk; to date, there have been few studies examining risk factors for aripiprazole-induced psychotic worsening. One study reported 40 % of patients worsened following a switch to aripiprazole, and patients already receiving relatively high antipsychotic doses may be at greater risk compared to those on lower doses (727 vs. 382 mg/day, respectively) (Takeuchi et al. 2009). A second study indicated that approximately 30 % of patients did not complete a switch to aripiprazole; descriptively, they received more typical antipsychotics and higher doses (≥1 DDD), were more ill, and had a longer course of illness compared to those who successfully completed the switch (Lin et al. 2009). Summarizing, it would seem that increased risk occurs in patients with chronic schizophrenia who are more ill and, as a result, are receiving relatively high antipsychotic doses. However, we are also reminded that these studies involved individuals undergoing a switch to aripiprazole, in contrast to reports where aripiprazole is added to existing treatment. Data drawn from switching studies must also factor in the possibility that the new drug, in this case aripiprazole, is not as effective as the antipsychotic the individual had been taking.

Differences were identified between cases where aripiprazole was added to current treatment (eight cases) and those where individuals were switched to aripiprazole from current treatment (14 cases). The main reason for introducing aripiprazole was efficacy (six cases) in the former, while efficacy (six cases) and side effects (six cases) were equally represented in the latter. In addition, number of cases where prior antipsychotic dose exceeded 1 DDD was slightly higher in cases where aripiprazole was added to existing antipsychotic treatment (six of eight cases) versus cases where individuals were being switched to aripiprazole (nine of 14 cases). Not surprisingly, it would appear that aripiprazole is added to current treatment in patients with treatment-resistant schizophrenia.

In terms of mechanism of action, exacerbation of psychotic symptoms by aripiprazole is most readily explained by its partial dopamine agonist properties (Burris et al. 2002); indeed, all 14 case reports that discuss the possible mechanism also implicate it. Notably, aripiprazole has a very high affinity for D2 receptors (Burris et al. 2002), in contrast to low affinities for other receptors other than 5-HT1A and 5-HT2A (DeLeon et al., 2004). That the cases occur in chronically treated individuals also calls into question the impact of long-term antipsychotic administration and dopamine upregulation (Silvestri et al. 2000), possibly resulting in increased sensitivity to dopaminergic perturbations (i.e., supersensitivity psychosis) as may occur with partial agonist activity via the addition of aripiprazole. To this point, 11 case reports posit a role for upregulation of postsynaptic dopamine receptors or supersensitivity psychosis (Moncrieff 2006).

The worsening of psychosis associated with aripiprazole, as detailed in these case reports, was not improved by altering the dose of aripiprazole, although resolution occurred frequently after aripiprazole discontinuation. This certainly offers at least indirect evidence supporting a causal relationship between aripiprazole and psychotic worsening.

Finally, from a clinical standpoint, it is also important to recognize that the symptoms related to the addition of aripiprazole were not confined to psychotic symptoms per se, but also included agitation, aggression, and activation. Indeed, cases have been reported where the central features were agitation (Cho and Lindenmayer 2009) or mania (Ducroix et al. 2008; Padala et al. 2007; Traber et al. 2007). Regarding mania induced by atypical antipsychotics, some reviews (Aubry et al. 2000; Benyamina and Samalin 2012; Michalopoulou and Lykouras 2006; Rachid et al. 2004) have suggested all types of atypical antipsychotics can induce mania or hypomania; thus, this phenomenon is not specific to aripiprazole. Notwithstanding, clinicians must also be vigilant of these symptoms and the possible need to consider aripiprazole discontinuation should they appear.

Unfortunately, a series of case reports does not allow us to establish the exact risk of aripiprazole for worsening psychotic symptoms, and predicting who is liable also remains an unanswered question. Any interpretation of the data is qualified by the diversity of the cases (e.g., illness severity, types and doses of prior antipsychotics, or concomitant medications) and variables beyond pharmacotherapy. For example, there is the possibility that factors such as natural course of illness or environmental stress play a role, although the quality of the causal relationship between aripiprazole and increased psychotic symptoms was evaluated according to the modified guidelines for evaluation of drug-associated events.

In conclusion, evidence suggests that at least a small number of patients with schizophrenia or schizoaffective disorder risk an exacerbation of psychotic symptoms if aripiprazole is added to existing antipsychotic treatment. Many of the cases reported involve patients who are quite ill and have been exposed to long-term antipsychotic treatment, but it may also be that this population is simply more likely to be exposed to augmentation strategies. There appears to be no relationship between its occurrence and age or gender; similarly, it can occur across all doses of aripiprazole and anytime from days to weeks after treatment commences. At this point, clinicians must simply be cognizant of the risk, vigilant of clinical worsening in the context of such a scenario, and aware that aripiprazole discontinuation appears the preferred strategy should it occur.