Abstract
Clozapine is an atypical antipsychotic used in treatment-resistant schizophrenia. Whilst clozapine is highly effective, there are some clinical scenarios, such as the emergence of severe side effects, that necessitate its discontinuation. There is an emerging literature suggesting that discontinuing antipsychotics, in particular clozapine, can cause an array of withdrawal symptoms. We review the evidence for the existence of clozapine-induced withdrawal symptoms, and in particular focus on withdrawal-associated psychosis, cholinergic rebound, catatonia and serotonergic discontinuation symptoms. To date, there has been surprisingly little clinical guidance on how to minimise the likeliness of withdrawal symptoms in patients who are stopped on clozapine abruptly or gradually. We discuss the key outstanding questions in this area and why there is a need for guidance on the management of withdrawal symptoms associated with clozapine discontinuation.
Introduction
Clozapine is an atypical antipsychotic that has been shown to reduce hospitalisation, mortality and risk of suicide in patients with treatment-resistant schizophrenia.1–3 Furthermore, studies – albeit mostly with relatively short follow up – suggest that clozapine is the only medication effective in treatment-resistant schizophrenia. 4 Clozapine is an antipsychotic of the dibenzodiazepine class and has a complex pharmacological profile. 5 It is a potent antagonist of H1, M1, 5-HT2B, 5-HT2C and α1 receptors; a potent inverse agonist of 5-HT2A receptors; and a relatively weak antagonist at D1, D2, D3 and D5 receptors, with higher affinity for D4 receptors.6,7
Whilst clozapine is predominantly reserved for patients with treatment-resistant schizophrenia, there are several clinical indications where there is a necessity to stop clozapine. Perhaps most notable amongst these are the emergence of potentially life-threatening adverse drug reactions, such as neuroleptic malignant syndrome, agranulocytosis and myocarditis. 8 Other indications for stopping clozapine include a lack of clinical efficacy, inadequate adherence to treatment or monitoring requirements and patient preference. 8 In some circumstances, clozapine has been discontinued successfully after symptom remission.9,10
Since the widespread use of clozapine, it has been recognised that stopping clozapine can lead to marked deterioration in clinical status in patients. Whilst this has been attributed largely to a relapse of the underlying mental disorder, it has become increasingly recognised that this may also be attributable to clozapine withdrawal symptoms. A wide spectrum of somatic and psychiatric symptoms have been reported (Figure 1), including psychotic, autonomic, gastrointestinal and psychomotor. 11 Some of these withdrawal symptoms, such as nausea and vomiting, have been described in other antipsychotics, whilst others, such as catatonia, appear to be specific to clozapine. 11 Whilst epidemiological evidence is limited, withdrawal symptoms have been reported most often in association with abrupt discontinuation. 11 Furthermore, symptoms often appear as a cluster, and elucidating these groups may yield insights into the underlying mechanisms responsible. To date, there has not been an attempt to summarise the evidence on clozapine-induced withdrawal symptoms. We sought to address this through a narrative review of the literature.

Clozapine withdrawal symptoms. A diverse range of symptoms have been reported; however, they can be broadly grouped according to the proposed underlying mechanism as (a) Withdrawal-associated psychosis, (b) Cholinergic discontinuation symptoms, (c) Serotonergic discontinuation symptoms and (d) Withdrawal-associated catatonia. Example symptoms within each group are listed. A patient may experience symptoms from one or more groups and symptom profile between groups may overlap.
Methods
The aim of this narrative review was to examine the evidence for specific groups of clozapine-induced withdrawal symptoms. To identify relevant studies, we searched the electronic database PubMed from inception to April 2020 using the search terms ‘clozapine’ AND ‘discontinuation’ OR ‘withdrawal’ OR ‘relapse’ OR ‘rapid onset psychosis’ OR ‘withdrawal-associated psychosis’ OR ‘dopamine supersensitivity’ OR ‘tardive psychosis’ OR ‘supersensitivity psychosis’. Eligibility required papers to be written in English and published in peer-reviewed journals reporting data on adult patients who were either decreased or discontinued on clozapine. Two researchers (GB and EO) performed the study search independently and in parallel. Where there were discrepancies, the researchers arrived at a consensus regarding eligibility. This was supplemented by searching the references of review articles on the topic. We then narratively summarised the evidence around the existence of distinct withdrawal symptom groups.
Withdrawal-associated psychosis
Clozapine discontinuation can precipitate the sudden emergence of psychotic symptoms that have been termed ‘withdrawal-associated psychosis’ or ‘supersensitivity psychosis’.12,13 This phenomena has been described in several antipsychotics, but is particularly associated with clozapine where it is estimated to occur in up to 20% of cases. 12 Early prospective studies on clozapine cessation suggest that withdrawal-associated psychosis typically occurs within 1 to 2 weeks of discontinuation14–16; however, there is evidence of an excess risk of relapse several months after clozapine and other antipsychotics are discontinued, suggesting that neural adaptations persist. 17
In a subgroup of patients, a ‘persistent post withdrawal psychosis’ has been described, whereby patients experience withdrawal-associated psychosis beyond the period typically associated with withdrawal symptoms. 18 Persistent post withdrawal psychosis is characterised by an exacerbation of psychotic symptoms extending beyond 6 weeks after a decrease, discontinuation or switch of antipsychotics. 13 Psychotic symptoms are often more severe than prior to treatment, may include new features and be less responsive to treatment.26,27
Withdrawal-associated psychosis is particularly related to the abrupt discontinuation of clozapine.14–16 Support for the existence of withdrawal-associated psychosis includes the evidence that symptoms are more severe than prior to starting clozapine in some patients.14,16 Furthermore, studies suggest that the dose required to achieve remission is often higher after restarting clozapine. 22 As well as occurring in the context of discontinuation and dose reduction, withdrawal-associated symptoms have also been suggested to occur between doses of clozapine,13,29 particularly in patients on a once daily regime. 30 With a mean half-life of 12 hours, clozapine plasma concentrations may fluctuate by over 50% within a patient on a stable dose; therefore, withdrawal symptoms between doses is plausible. Whilst further empirical research in this area is indicated, preliminary evidence supports this assertion. 26
Patients with treatment-resistant schizophrenia are particularly vulnerable to symptom deterioration after clozapine discontinuation. 28 Several studies support clozapine’s superiority over other antipsychotics, albeit with mostly short periods of follow-up.4,29 In contrast, there is a paucity of high-quality evidence exploring the most effective antipsychotic after clozapine discontinuation. No head-to-head studies have been reported, preventing direct comparisons. One randomised clinical trial found that switching from clozapine to the atypical antipsychotic zotepine led to a significant worsening of symptoms compared with clozapine continuation. 30 The only randomised placebo controlled trial to date found that switching from clozapine to olanzapine led to a reduced likeliness of withdrawal-associated psychosis following abrupt discontinuation, compared with placebo. 31
The pathophysiology of withdrawal-associated psychosis is not fully understood, and there is debate as to whether it reflects a relapse of the underlying psychotic disorder, or a distinct clinical phenomenon. Distinguishing between the two is challenging as most patients treated with clozapine have a psychotic disorder. However, there are case reports of withdrawal-associated psychosis occurring in patients discontinued on antipsychotics without a prior history of psychosis, suggesting that, at least in some patients, withdrawal-associated psychosis may represent a distinct clinical phenomenon.32,33 Meta-analytic evidence indicates that relapse rates in the 6 months following abrupt discontinuation of antipsychotics are significantly higher in patients with schizophrenia than would be predicted based on the natural course of the disorder. 17 Furthermore, the abruptness of onset and the associated increased severity of symptoms (in some cases) are in keeping with the suggestion that it is distinct from a relapse of the underlying psychotic illness (Table 1). 12
Withdrawal-associated psychosis studies. .
ECT, electroconvulsive therapy; EPSE, extrapyramidal side-effects; RCT, randomised controlled trial.
Cholinergic symptoms
Cholinergic discontinuation symptoms, also known as ‘cholinergic rebound’, are characterised by a range of psychiatric and somatic clinical features including nausea, vomiting, confusion, insomnia and dystonia, 35 36 which are thought to arise as a result of overactivity of the cholinergic system (Table 2). Several case reports have described the onset of cholinergic symptoms following clozapine discontinuation.31,34,35,37–41 These cases suggest that symptoms appear within a few days and continue for several weeks34,42–45; however studies in other antipsychotics suggest that symptoms may continue for longer. 11 Furthermore, cholinergic rebound has been reported in patients on a range of doses of clozapine, including doses as low as 50 mg per day. 37
Cholinergic withdrawal symptom studies.
OCD, obsessive compulsive disorder.
Cholinergic rebound is not isolated to antipsychotics and has been associated with the discontinuation of a range of central nervous system medications, including tricyclic antidepressants and anti-parkinsonian drugs.36,42,46–50 Common to all the drugs associated with cholinergic rebound is a close affinity toward cholinergic receptors. The most likely mechanism underlying cholinergic rebound is the upregulation of muscarinic acetylcholine receptors due to prolonged exposure of clozapine, leading to super-sensitivity after discontinuation (Table 2).36,51
Catatonia
Catatonia is a psychomotor disorder characterised by a range of psychomotor features including stupor, posturing and echo phenomena. 52 Catatonia is associated with a range of psychiatric disorders, such as depression and schizophrenia, as well as non-psychiatric disorders such as autoimmune encephalitis and withdrawal states. Several case reports have described the acute onset of catatonia following the withdrawal of clozapine (Table 3). This has typically been reported in the context of stopping clozapine abruptly.
Withdrawal catatonia studies.
ECT, electroconvulsive therapy.
The prevalence of clozapine-withdrawal-induced catatonia is unknown; however, a recent systematic review identified 20 reported cases in the literature. 59 It has been observed predominantly in patients treated with clozapine for several years and without a history of catatonia. Symptoms generally emerge within a week of clozapine discontinuation. 59 Interestingly, catatonia is not associated with the discontinuation of other antipsychotics, which may suggest that the unique pharmacology of clozapine plays an important role. In contrast, withdrawal catatonia has been shown to occur with other medications, most notably benzodiazepines. The mechanisms underlying clozapine-withdrawal-induced catatonia are not fully understood; however, hypoactivity of the GABAergic system is strongly implicated in the emergence of catatonia. Clozapine does not have a direct effect upon the GABA system (unlike benzodiazepines); however, prolonged use has been shown to lead to GABA receptor adaptation and a reduction in GABAergic effects (Table 3). 60
Serotonergic symptoms
Serotoninergic discontinuation symptoms (Table 4) including agitation, diaphoresis, clonus and hyperreflexia have been described in a small group of patients following the cessation of clozapine61,62; however, the incidence is unknown. Case report evidence suggests that clozapine-induced serotonergic symptoms can arise in the presence or absence of concomitant serotonergic medications.61–63 Symptoms have been described as typically appearing within a few days of clozapine discontinuation. 62
Serotonin withdrawal symptom studies.
Serotonergic discontinuation symptoms are not unique to clozapine and have been described in other antipsychotics, particularly those with 5-HT2A antagonistic properties, such as aripiprazole and quetiapine.64,65 However, serotonergic discontinuation symptoms appear to be associated more closely with clozapine, which may reflect the degree of affinity toward serotonergic receptors. The mechanism by which clozapine discontinuation induces these symptoms is not clear; however, it has been postulated to relate to clozapine’s direct effect on serotonergic receptors. Clozapine is a potent 5-HT2A antagonist and prolonged use may lead to an upregulation of serotonin receptors, resulting in super-sensitivity (Table 4). 62
Discussion
A review of the literature suggests that clozapine discontinuation is associated with an array of withdrawal symptoms with important clinical implications. There is evidence for the existence of four groups of withdrawal symptoms: withdrawal-associated psychosis, cholinergic rebound, catatonia and serotonergic discontinuation symptoms. Furthermore, there is some evidence to suggest that these symptom groups are underpinned by distinct neural mechanisms. Clozapine has a complex pharmacological profile and induces a range of neuronal changes at the receptor level, especially when taken over an extended period. Evidence suggests that these neural adaptations underlie the emergence of withdrawal symptoms upon the discontinuation of clozapine.
What are the clinical implications of these findings? To date, there has been little guidance on how to safely discontinue clozapine to reduce the likeliness of withdrawal symptoms. This is likely to reflect, at least in part, a lack of awareness of this important topic alongside a sparse empirical basis. Clozapine discontinuation rates are estimated to be as high as 45% after 2 years. 66 As such, there is a need for evidence-based guidelines on the prevention and management of withdrawal syndromes associated with clozapine discontinuation.67,68 For example, in light of the evidence that withdrawal symptoms are likely to be underpinned by neuronal adaptions giving rise to receptor super-sensitivity, when stopping clozapine, a slow taper (for example, over several months or years) may help to reduce the risk of withdrawal symptoms. 69
Guidelines would have at least two clear patient benefits. First, they would minimise the potential harms associated with clozapine withdrawal where discontinuation is indicated. Second, they may help to address clinicians’ reservations around offering clozapine to patients due to the risks associated with clozapine discontinuation. As a consequence, this may help reduce the underutilisation of clozapine in patients who may benefit.
Although withdrawal symptoms as a result of clozapine discontinuation have been described for over three decades, 91 there remains a general lack of awareness of this phenomenon. This may result in clinicians misdiagnosing withdrawal symptoms for relapse of the underlying disorder. 92 A potential consequence of this conflation is that it may lead to the conclusion that all patients should be maintained on clozapine indefinitely to prevent relapse. For example, it is plausible that some of the negative outcomes attributable to stopping clozapine1–3 are due to withdrawal-associated symptoms. Consequently, it may be that (at least in some patients) outcomes could be improved by simply optimising the discontinuation of clozapine to minimise the risk of withdrawal symptoms.
Further research is vital to better prevent, detect and treat clozapine withdrawal. Whilst there is evidence for withdrawal-associated psychosis, catatonia, cholinergic and serotonergic symptoms occurring as a result of clozapine cessation, there is a need for large prospective studies to further examine these withdrawal symptoms using standardised measures to address the many important outstanding questions. For example, from an epidemiological perspective, the prevalence of clozapine discontinuation withdrawal symptoms remains unknown. Risk factors for developing clozapine-withdrawal symptoms also remain largely unexplored. Whilst immediate cessation of clozapine is strongly associated with the emergence of withdrawal symptoms, to what extent other variables (such as clozapine dose, treatment duration and demographic variables) are important remains to be addressed. From a mechanistic perspective, it is not clear to what extent clozapine-induced withdrawal effects, such as withdrawal-associated psychosis and cholinergic rebound, overlap. From a preventative approach, the optimal tapering regime to minimise the risk of withdrawal symptoms has also not been explored empirically; however, studies are underway. 93 Whilst a patient-rated scale to measure the success of antipsychotic and antidepressant discontinuation has recently been developed, 94 the scale does not differentiate between the withdrawal symptoms described in this review, which may also be present with other psychotropics. 11 Finally, the role of medication to prevent and treat clozapine-withdrawal symptom is largely unknown. Such advances could greatly benefit patients at risk of developing withdrawal symptoms. Furthermore, the availability of such treatments would likely have the additional advantage of lowering the threshold for clinicians to trial clozapine in patients who may benefit.
