FormalPara Key Points

Clozapine discontinuation after myocarditis typically has a detrimental impact on patients’ long-term outcome, but converging evidence indicates that a re-challenge is a safe and feasible approach in many cases after at least three months of recovery without clozapine.

CRP is an essential screening parameter during both standard clozapine titrations and during re-challenges after myocarditis because it is sensitive to both clozapine-induced myocarditis and other clozapine-induced inflammatory adverse events.

Current evidence indicates that slow dose titration is the most promising approach to prevent adverse cardiac events during clozapine treatment.

Very slow dose titration is also the most important measure to increase cardiac safety during re-challenges after myocarditis in combination with a stringent screening protocol.

1 Introduction

Clozapine is a vital treatment option for patients with a poor or treatment-resistant course of schizophrenia and currently remains the most effective and the only approved drug for treatment-resistant schizophrenia (TRS) [1,2,3,4]. Its relatively high side-effect load makes the use of clozapine comparatively demanding for prescribers and requires comprehensive side-effect detection and management [5]. Nevertheless, due to its unique efficacy, concerted efforts to address the considerable underutilization of clozapine across industrialized countries [6] are imperative [7].

Clozapine-induced myocarditis (CIM) is an important serious adverse event whose detection and management poses unique challenges. Moreover, it can limit the use of clozapine for individuals who have not experienced significant improvement with other antipsychotic compounds. Therefore, strategies to prevent CIM, to increase patient safety in cases of CIM and to avoid permanent discontinuation of clozapine after CIM are urgently needed. The aim of this narrative review is to summarize relevant aspects of CIM including epidemiology, pathophysiology, risk factors and clinical presentation as well as CIM prevention, detection and management strategies. Furthermore, we provide a critical appraisal and synthesis of recommendations regarding the rationale and procedures for post-CIM re-challenges. We also discuss how both laboratory and imaging biomarkers can aid the timely detection of CIM and increase the chances for successful subsequent clozapine re-challenges.

While we focus primarily on the current state of the art regarding CIM management and post-CIM re-challenges, we also discuss the implications of widely different practices regarding clozapine use across psychiatric health care systems including in non-industrialized countries with severely limited resources. To address these discrepancies, we provide recommendations for both minimal and optimal CIM screening and diagnostic procedures.

2 Literature Search

We based our review on a MEDLINE and Google Scholar search for all publications primarily addressing CIM and re-challenges after CIM. We included relevant clinical studies as well as meta-analyses and reviews. We selected all articles that were published until 10th May 2024. We searched for publications containing the following Medical Subject Headings (MeSH) terms: treatment-resistant schizophrenia [AND] criteria, clozapine [AND] xx, with xx reflecting the topic we aimed to focus at, at that time point (e.g. children and adolescents, shared decision-making, guidelines, myocarditis, cardiomyopathy, re-challenge [AND] cardiomyopathy, re-challenge [AND] myocarditis, myocarditis [AND] pathophysiology, vaccination, Coronavirus disease 2019, mRNA vaccines, pharmacokinetics, valproate, olanzapine, quetiapine). Furthermore, we manually screened reference lists of topical review articles. In cases where we did not have any access to the full article, we contacted the study authors. We only included articles for which we received the full text.

3 Clozapine-Induced Myocarditis

3.1 Definition and Epidemiology

CIM is defined as myocardial inflammation following clozapine exposure [8,9,10]. According to a recent meta-analysis encompassing more than 250,000 participants, the mortality rate attributable to CIM is about 4 per 10,000 individuals per year [11]. Currently, we lack a clear picture of the full spectrum of long-term consequences of undetected CIM. Specifically, it remains unclear whether all undetected cases have a lethal course or whether a relatively benign and self-remitting course is also possible.

The risk for CIM is highest during the first eight weeks after treatment initiation [10,11,12,13,14,15,16,17,18]. Notably, there have also been reports of a delayed onset of CIM after long-term treatment [16, 19, 20] but, even more importantly, of cases during re-exposure following a first successful trial [21, 22]. Undetected cases of CIM during the initial trial could also account for some of these observations [22, 23].

Myocarditis ranks among the most relevant and consequential clozapine-associated adverse events [24, 25]. Compared with clozapine-induced agranulocytosis (CIA) and neutropenia, incidence rates of CIM are generally higher, with estimates ranging from 0.03% to 8.5% [21, 25,26,27]. The exact causes for this considerable variability across countries remain unclear. Interestingly, the highest rates have been reported in Australia, and lowest in Denmark and the Netherlands [11, 28]. Stringent monitoring and reporting regimes, which should increase the detection and report of less obvious cases, as well as comparatively fast clozapine titration rates might account for the high rates described in Australia [10, 13, 14, 29, 30]. Importantly, registry-based data reporting CIM prevalence in Denmark only encompass outpatients with initially slow titration regimes [31, 32]. Dutch titration regimes in both out- and inpatient settings are similarly slow [13, 31, 33, 34].

3.2 Pathophysiology

Overall, the limited published findings [25, 35] indicate a heterogeneous pathophysiology of CIM. To date, the best evidence implies an immunological pathomechanism [10, 36]. CIM is typically accompanied by peripheral eosinophilia and eosinophilic myocardial infiltrates, which implicates a type I immunoglobulin E-mediated hypersensitivity reaction in its pathophysiology [12, 35, 37,38,39,40,41]. A hypereosinophilic syndrome affecting myocardial tissue but potentially also other internal organs, including the colon, kidney and pancreas, has been proposed as another possible cause for CIM [39]. This could explain the lack of other immunoglobulin-E-associated symptoms such as urticaria [39]. However, peripheral eosinophilia is not consistently observed in CIM, making a hypereosinophilic syndrome as a central cause less likely [12]. Moreover, differentiating between these mechanisms might be difficult since type I immunoglobulin E-mediated hypersensitivity is also associated with increased eosinophil levels [39, 41].

Increasing evidence for the existence of a ‘clozapine storm’ reaction [10] comprising transient fever and elevated cytokine levels during the first few weeks of clozapine treatment suggests the involvement of clozapine-induced pro-inflammatory cytokine release in the pathophysiology of CIM [40, 42, 43]. Importantly, this systemic inflammatory response can induce both myocarditis-like symptoms as well as actual myocardial inflammation [10, 36, 44].

Moreover, a hypersensitivity reaction comparable to lamotrigine-induced Stevens–Johnson syndrome (SJS) and similarly triggered by rapid dose titration has also been proposed as a potential pathophysiological mechanism [14, 18, 44,45,46,47]. SJS is a delayed-type hypersensitivity reaction commonly caused by treatment with lamotrigine, carbamazepine, allopurinol, sulphonamide antibiotics and nevirapine [48, 49], manifesting within the first two months of treatment. For lamotrigine, there is a clear association between severe hypersensitivity reactions and rapid dose increases [50]. Importantly, the introduction of an obligatory gradual lamotrigine titration strategy in 1994 resulted in a ten-fold decrease in the incidence of SJS from 1% to less than 0.1% [51,52,53,54]. The potential pathophysiological overlap between CIM and SJS implies that slow titration rates are crucial for preventing CIM and for successful re-challenges after CIM [45]. Notably, the drug reaction with eosinophilia and systemic symptoms (DRESS) syndrome, a rare, severe adverse event of clozapine featuring widespread inflammation including myocarditis, also appears to result from a hypersensitivity reaction [44]. This potential pathophysiological overlap with CIM suggests the existence of a continuum of clozapine-induced inflammatory adverse events, which we will discuss in greater detail in a subsequent section.

Additionally, a hypercatecholaminergic state resulting in myocardial inflammation [55] as well as increased myocardial oxidative stress and downregulation of endogenous antioxidants [35] have been observed in animal models of CIM. These studies demonstrated a clear correlation with the clozapine dose. Furthermore, there is recent evidence indicating that a relatively rapid increase in the plasma concentration of clozapine-N-oxide as well as a high ratio of clozapine-N-oxide relative to N-desmethyl-clozapine increase the risk for CIM [56]. N-oxidation of clozapine is particularly pronounced at high plasma clozapine levels when N-desmethylation saturates. N-oxidation of clozapine leads to an increased formation of unstable reactive nitrogen species, i.e. nitrenium and iminium intermediates, which may be involved in clozapine-induced cardiotoxicity [56, 57].

3.3 Risk Factors

Despite the accumulating evidence for CIM risk factors (Table 1), reliable predictors for an individual’s risk for CIM are currently lacking. In general, particular caution is warranted in cases with a history of cardiac disorders and myocarditis [58, 59]. Data from clinical studies and animal models clearly indicate that rapid dose titration constitutes an important risk factor which can trigger the immunological consequences outlined above [25, 35, 55, 60,61,62]. This implies that people with a slow metabolizer status for clozapine are predisposed for developing CIM, if titration rates are not adjusted accordingly [35, 60,61,62,63,64]. Here, ethnic ancestry, as defined by DNA ancestry group, also needs to be considered, given that it has an important influence on clozapine metabolism [44]. Patients of Asian descent appear to respond to doses, which are approximately 35–40% lower compared with patients of European descent [31]. The slower clozapine metabolism in people with Asian ancestry could partly account for the higher frequency and severity of CIM observed in Asian countries [18, 65, 66].

Table 1 CIM risk factors
Full size table

A recent meta-analysis identified concurrent valproate treatment as the clearest risk factor for CIM [25]. Pharmacokinetic interactions—namely valproate-mediated slow clozapine metabolism—have been discussed as one potential underlying mechanism [25, 61, 67,68,69,70,71,72]. Furthermore, competitive protein binding has also been implicated as a cause of valproate-mediated clozapine potentiation [61, 73], especially in carriers of specific uridine 5′-diphospho-glucuronosyltransferase polymorphisms [68]. Overall, these findings indicate that concurrent valproate treatment raises the risk for CIM predominantly by increasing plasma clozapine levels, mirroring the effects of rapid dose titrations. Consequently, although pharmacokinetic in vivo studies and larger prospective studies are still lacking and current clinical findings remain somewhat conflicting [68, 73, 74], concurrent administration of valproate should be avoided, especially during the initial clozapine titration [62, 67].

Pharmacokinetic interactions which reduce clozapine metabolism might also play a role for concurrent treatment with olanzapine [18, 27, 62, 75]. On the one hand, olanzapine monotherapy is associated with an elevated risk for myocarditis [14, 41, 66, 76, 77]. However, this risk is lower compared with clozapine [41, 76]. On the other hand, cytochrome P450 (CYP) 1A2 enzyme saturation resulting from a combined use of olanzapine and clozapine has been implicated in elevated CIM risk [14, 18, 41, 76, 78]. Given that both clozapine and olanzapine are substrates of the CYP1A2 enzyme [79, 80], such a pharmacokinetic interaction is conceivable. It is further supported by the fact that both compounds are impacted by polycyclic aromatic hydrocarbons contained in cigarette smoke—a strong inducer of CYP1A2 [81, 82]. These findings also imply that non-smokers, who are a minority among patients diagnosed with schizophrenia [83] and have a comparatively slow clozapine metabolism, are at a relatively greater risk for CIM [62].

Obesity, which can alter clozapine distribution and reduce its elimination, constitutes another potential risk factor [84,85,86], which also necessitates adequately adjusted titration regimes [62]. Higher age has also been implicated as a risk factor [25, 62, 87], but current findings remain conflicting. This is reflected by reports of higher CIM rates in young male individuals aged between 20 and 40 years [25, 87, 88]. Available pharmacokinetics data for children and adolescents treated with clozapine, while limited, indicate no clear differences regarding clozapine metabolism [89]. Sex has a clear impact on clozapine metabolism, with females having lower CYP1A2 and CYP2C19 enzyme activity [62, 90]. However, data regarding sex as a risk factor for CIM remain inconclusive [25, 62, 88].

Human in vivo and in vitro studies have demonstrated immunomodulatory effects of clozapine, which include elevated cytokine levels, e.g. for C-reactive protein (CRP), tumor necrosis factor alpha, interleukin-(IL)-6 and soluble IL2 receptor, particularly during the first 2 months of treatment [8, 36, 91, 92]. The initial increase of these markers could contribute to inflammation in several tissues. Furthermore, CRP and cytokines inhibit CYP1A2 enzyme activity and increase plasma clozapine levels [36, 62, 91, 93,94,95]. These mechanisms are also very relevant in the context of messenger ribonucleic acid (mRNA) vaccines for severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2). These vaccines are associated with a generally increased risk for myocarditis [93, 95,96,97,98,99,100,101,102]. Furthermore, a systemic inflammatory response following mRNA vaccinations with elevated cytokine levels and subsequently elevated plasma clozapine levels can specifically increase the risk for CIM during the initial stages of clozapine treatment [36, 62, 103]. Yet, currently there is no clear evidence for increased clozapine adverse event rates after coronavirus disease 2019 (COVID-19) vaccination [104]. Consequently, there is a clear consensus that the benefits of SARS-CoV-2 vaccination outweigh the risk for developing myocarditis [102]. This assessment is in no small part based on the finding that COVID-19 infections can also cause (peri-)myocarditis [96, 98, 102]. Furthermore, COVID-19 infections can also increase plasma clozapine levels to a considerable degree via cytokine-mediated decrease of CYP1A2 enzyme activity [95]. Most importantly, COVID-19-related mortality in patients diagnosed with schizophrenia is increased substantially [105], underscoring that psychiatrists should ensure sufficient vaccination status of their patients.

Concurrent treatment with quetiapine has also been discussed as CIM risk factor. It has been suggested that quetiapine might have additive pharmacodynamic effects in the context of clozapine-induced inflammation [41, 66, 77]. While a recent meta-analysis did not implicate quetiapine as a CIM risk factor [25], this might be attributable to a lack of power and the considerable heterogeneity of the analysed studies.

As outlined above, reports of CIM in children and adolescents remain very limited [106]. Available data indicate no relevant differences regarding clinical symptoms and onset [106] and provide no evidence for specific CIM risk factors in this particular patient group. In addition to concurrent treatment with olanzapine or valproate, concurrent treatment with quetiapine also emerged as a risk factor for a particularly serious course of CIM in children and adolescents [106]. Notably, this risk was highest for quetiapine. While the underlying mechanisms remain unclear, there is evidence for an increased myocarditis risk arising from supratherapeutic quetiapine doses and rapid quetiapine titration [14, 78, 106].

While genetic risk factors for CIM are likely given the findings for clozapine-induced agranulocytosis (CIA) [107], thus far no clear genetic markers have been identified [25, 108]. There is preliminary evidence implicating human leukocyte antigen polymorphisms [65, 109]. Furthermore, polygenic risk scores for schizophrenia have been associated with cardiac phenotypes that increase the risk for cardiac diseases [110]. This also raises the question whether there might be a link between CIM and schizophrenia polygenic risk. Presently, to our knowledge there is no evidence for genetic variants—especially in the CYP1A2 gene—associated with a slow metabolizer status for clozapine, which would constitute an obvious genetic risk factor given the evidence reviewed above [41, 109]. Potential candidates include rare, high-effect-size variants, large structural variants or functional haplotypes in genes relevant to clozapine metabolism, which are not detectable by currently available methods [109]. Furthermore, phenoconversion—a mismatch between the genotype-predicted CYP450-enzyme metabolizer status and the actual capacity to metabolize pharmacological compounds—could also account for a relevant part of the variability in clozapine metabolism [64, 111, 112]. Phenoconversion arises from the nongenetic factors influencing clozapine metabolism reviewed above, i.e. somatic comorbidities, concomitant medication, age, smoking and inflammation [64, 111, 112]. Consequently, in cases of an apparent poor metabolizer status phenoconversion always needs to be considered and addressed [113].

3.4 CIM and the Spectrum of Clozapine-Induced Inflammatory Events

There is increasing evidence that clozapine treatment initiation is associated with proinflammatory responses. Pre-clinical in vitro and in vivo data indicates that clozapine causes an initial IL-1b release through inflammasome-dependent caspase-1 activation [114], which could provide a parsimonious account for these phenomena. However, due to a lack of systematic studies, it remains unclear whether there are common pathophysiological mechanisms underlying the different manifestations of clozapine-induced inflammation [44]. Clozapine-induced inflammation can affect several organs and result in a variety of clinical presentations, including but not limited to myocarditis, pericarditis without myocarditis, pancreatitis, nephritis, hepatitis, pneumonitis, serositis [36, 44, 114, 115] and DRESS syndrome [116, 117]. Hence, CIM appears to be the most prominent outcome among a spectrum of clozapine-induced organ inflammation. Notably, CIA has also been linked to inflammation [114], which raises the possibility that it might at least partly belong to the same spectrum of adverse events.

In the context of CIM, clozapine-induced pericarditis is the most pertinent clozapine-induced inflammatory event. Its clinical presentation might mimic CIM symptoms, but it remains less diagnosed probably due to lack of awareness [118]. Moreover, the predictive value of cardiac biomarkers for pericarditis is less well established [118, 119], given that elevated troponin levels appear to be primarily associated with myocardial tissue damage [119]. However, clozapine-induced pericarditis is also associated with elevated CRP levels. Consequently, rising CRP levels without elevated troponin during clozapine titration should still prompt the full cardiological diagnostic workup regarding a potential pericarditis as part of a general search for clozapine-induced inflammation. Moreover, any manifest organ inflammation warrants particular caution to prevent further progression including an affection of the myocardium [120].

This approach is also crucial for detecting and preventing DRESS syndrome, a rare adverse event which constitutes the most serious manifestation of clozapine-induced inflammation. It likely results from a hypersensitivity reaction and can lead to potentially life-threatening multi-organ inflammation, potentially affecting the peri- and myocardium, the kidneys, pancreas, lung, liver, thyroid, small intestine and colon [116, 121,122,123]. DRESS syndrome typically presents with rash, fever and lymphadenopathy, as well as eosinophilia, lymphocytosis and thrombopenia [116]. An onset during the first weeks after clozapine initiation is typical, mirroring CIM and other mono-organ inflammations [64, 122]. Concurrent treatment with valproate or lithium appears to increase the risk for DRESS syndrome [64, 122].

The growing body of evidence for the considerable extent of the immunomodulatory effects of clozapine underscores the need for pharmacovigilance beyond CIM. One important clinical implication is the special relevance of CRP as a marker for the full spectrum of associated adverse events. Moreover, there is now tentative evidence indicating that the immunomodulatory impact of clozapine is not limited to its crucial role in inducing inflammatory adverse events. Rather, the observation of a correlation between reduced immunoglobulin levels (A and G) and psychopathological improvements over twelve weeks of clozapine treatment suggests that immunomodulation might also partly underlie its therapeutic effects [124].

4 Clinical Aspects of CIM

4.1 Clinical Presentation

Typically, CIM occurs during the first eight weeks of clozapine treatment [10, 15, 16, 19, 25, 26]. Clinical presentation of CIM is often unspecific (Fig. 1). Patients may report flu-like and non-specific symptoms such as fever, dyspnoea, myalgia, fatigue and malaise apparently arising from the clozapine-induced systemic inflammatory response [10, 26, 125]. More specific symptoms include chest pain, hypotension, palpitation, tachycardia and peripheral oedema [10, 26, 125]. Importantly, symptoms reflecting cardiac affliction may be largely absent. Rather, patients might present with gastrointestinal and urogenital disturbances including nausea, vomiting, diarrhoea and dysuria [10, 125,126,127]. CIM can be associated with tachycardia [16, 127] or an increase in heart rate of approximately 30 beats per minute above baseline [125]. This can be accompanied by a drop of systolic blood pressure. Importantly, isolated tachycardia and orthostatic dysregulation are common side effects of clozapine [10, 125, 128, 129]. The prevalence of tachycardia without further signs for CIM in patients with clozapine ranges from 3% to 67% [128, 129]. Because both tachycardia and hypotension are of low diagnostic value, additional parameters in support of a CIM diagnosis are required to avoid premature clozapine termination [10].

Fig. 1
figure 1

CIM clinical presentation. Patients with CIM report unspecific clinical symptoms as CIM can affect several organs including cardiovascular system, gastrointestinal and urogenital tract. Clinical findings are accompanied by electrophysiological and laboratory abnormalities which include increased troponin and CRP as well as BNP and NT-proBNP. ECG may encompass ambiguous changes. TTE findings can include left ventricular impairment and pericardial effusion. CMR can serve as a more precise approach to detect myocardial inflammation. EMB allows to rule out viral myocarditis. BNP brain natriuretic peptide, CIM clozapine-induced myocarditis, CMR cardiac magnetic resonance imaging, CRP C-reactive protein, ECG electrocardiogram, EMB endomyocardial biopsy, NT-proBNP N-terminal-pro hormone and brain natriuretic peptide, TTE transthoracic echocardiogram. Adapted from ‘Human Internal Organs’, by BioRender.com (2023). Retrieved from https://app.biorender.com/biorender-templates

Full size image

This also applies to clozapine-induced fever, which manifests without concomitant infection [130]. Onset of clozapine-induced fever typically occurs within 5–15 days after treatment initiation. Fever typically lasts for about 3–5 days and often resolves spontaneously [130]. It has been suggested that clozapine-induced fever could signify a clozapine-induced cytokine release triggered by a hypersensitivity reaction during rapid titration [44]. Importantly, this inflammatory response might merely mimic symptoms of myocarditis in the absence of myocardial inflammation. Yet, as outlined above, clozapine-induced cytokine release can lead to myocarditis, serositis, pneumonitis, hepatitis, pancreatitis or nephritis, among others [36, 44].

Consequently, fever during the initial stages of clozapine treatment should prompt a search for inflammation affecting one or more of these organ systems. Because fever could also indicate clozapine-induced neutropenia and agranulocytosis, such a diagnostic workup should always include a white blood cell differential [10, 16]. Importantly, the clinical presentation of CIM and pericarditis overlap considerably. Moreover, CIM can also imitate the initial stages of DRESS syndrome, which constitutes an important albeit rare differential diagnosis [117, 120].

4.2 CIM Screening

Frequent screening for signs of CIM are an important element of clozapine treatment initiation (Fig. 2) [127, 131]. To this end, clinicians should not rely solely on the unspecific clinical presentation. Rather, focusing on laboratory, electrophysiological and cardiac imaging indicators of CIM is essential. This strategy allows for a timely detection and management of CIM, considerably enhancing patient safety [10]. Among routine assessments, laboratory abnormalities are the most consistent and reliable indicators of CIM (Fig. 2). Suitable laboratory screening parameters for CIM comprise troponin, CRP, creatine kinase (CK) and the myocardium-specific isoenzyme muscle–brain type CK (CK-MB) as well as eosinophils [10, 11, 15, 125, 132]. These tests should be performed at baseline and weekly for the first 8 weeks [15]. While many authors consider this to be a mandatory strategy, there is currently no clear consensus among experts regarding laboratory screening for CIM [133, 134].

Fig. 2
figure 2

Optimal CIM monitoring and management. Optimal cardiovascular monitoring includes frequent screening for clinical, electrophysiological and laboratory signs at baseline and during the first 8 weeks. Laboratory screening parameters for CIM comprise CRP and troponin, CK and the myocardium-specific isoenzyme CK-MB. Adding a screening of BNP (or NT-proBNP) at least at baseline and in case of suspected CIM allows detecting CIM at early stages. Electrophysiological monitoring includes weekly ECG, daily measurements of heart rate, blood pressure and body temperature. Cardiovascular imaging includes TTE (at baseline and in case of suspected CIM) and if available CMR for confirmation of suspected CIM. EMB should be considered in case of suspected viral myocarditis. Commonly CIM diagnosis is based on laboratory parameters (i.e. CRP and troponin). CRP increases with blood levels that fall in the range of 50–100 mg/L and troponin increases with blood levels that fall in the range of < 2× ULN necessitate intensified monitoring (daily laboratory assessment; ECG and TTE) and a search for alternative causes. Increases in CRP blood levels exceeding the safety threshold of > 100 mg/L and increasing troponin blood levels exceeding the threshold of 2× ULN necessitate termination of clozapine, daily monitoring, initiation of a cardioprotective treatment and a search for an alternative cause. Confirmed CIM also requires avoiding physical exercise for at least 3 months. BNP brain natriuretic peptide, BP blood pressure, CIM clozapine-induced myocarditis, CK creatine kinase, CK-MB muscle–brain type CK, CMR cardiac magnetic resonance imaging, CRP C-reactive protein, ECG electrocardiogram, ECT electroconvulsive therapy, EMB endomyocardial biopsy, HR heart rate, NT-proBNP N-terminal-pro hormone and brain natriuretic peptide, TTE transthoracic echocardiogram, ULN upper limit of normal. Created with BioRender.com

Full size image

In a recent Treatment Response and Resistance in Psychosis (TRRIP) consensus paper, CRP and troponin T achieved only a guidance-level recommendation, e.g. an expert agreement proportion of above 50% [133]. Here, CK and CK-MB were not considered, most likely because of their more limited sensitivity for detecting myocardial damage [63, 135]. Conversely, in a large health register analysis, troponin T and CRP showed excellent diagnostic value [10]. Notably, there is some evidence that in the majority of CIM cases fever or increased CRP levels precede elevated troponin levels by several days [136, 137]. Both parameters are used in a currently ongoing multisite investigation of clozapine as a second-line treatment in schizophrenia [138], underscoring their relevance for establishing myocarditis as a safety endpoint in clozapine trials.

Assessing brain natriuretic peptide (BNP) or N-terminal prohormone of brain natriuretic peptide (NT-proBNP) at baseline and in case of suspected CIM is also recommended [125, 139,140,141]. Notably, several studies indicate that, compared with BNP, NT-proBNP levels might increase already at an earlier stage of myocarditis [139, 142]. Importantly, NT-proBNP baseline assessments are part of the Maudsley Treatment Review and Assistance Team (TREAT) recommendations, even though these are geared towards outpatients using a slow titration scheme of approximately 8 mg per day [143].

Recommended cardiological monitoring also includes, daily measurements of heart rate, blood pressure and body temperature as well as weekly electrocardiography (ECG) [16, 131]. The inclusion of ECG in CIM screening procedures [16] is justified by its high sensitivity for detecting arrythmia—an important complication of myocarditis with potentially severe consequences [144]. If available, transthoracic echocardiogram (TTE) can be performed prior to clozapine treatment initiation and in case of suspected CIM [10, 76, 145]. Such an approach can help to differentiate between pre-existing cardiac conditions and cardiac abnormalities caused by clozapine initiation. This might also alleviate patients’ concerns regarding their cardiac health and the safety of clozapine treatment. However, TTE is not a mandatory part of CIM screening, but should be performed under optimal conditions.

The combined use of these measures increases the initial treatment burden and might discourage prescribers and patients from a clozapine trial. Nevertheless, in our view the limited available evidence regarding CIM risk factors currently precludes a more targeted approach. Due to their very low specificity, relying solely on vital sign screening and clinical symptoms carries the risk of a delayed CIM diagnosis. Furthermore, the considerable psychopathology of many patients requiring clozapine likely limits their ability to timely report clinical signs of myocarditis.

Considering the full spectrum of potential clozapine-induced inflammation reviewed above, conducting a weekly CRP screening is also crucial for the timely detection of clozapine-induced inflammation affecting other organ systems. Given the vastly different conditions across health care systems in terms of available resources, proposing a minimal sets of screening procedures for low-resource conditions and an optimal set for conditions in which limited resources are not an issue might be most appropriate. The minimal set of parameters should be used primarily in health care settings faced with a lack of resources. Conversely, the optimal set of parameters, which emphasize maximal patient safety, should be used whenever possible. Our CIM screening recommendations (Table 2) are based on the evidence and considerations discussed above.

Table 2 Proposed minimal and optimal parameters for CIM screening and diagnostic workupa
Full size table

4.3 Diagnostic Workup

CIM diagnosis is primarily based on CRP and troponin levels (Fig. 2). To this end, an alert and a safety threshold have been established. The alert threshold is defined as either CRP increases within the range of 50–100 mg/L or troponin increases to less than twice of the upper limit of normal (ULN). The alert threshold requires intensified monitoring, a search for alternative causes of suspected myocarditis [15, 127] and precludes any further clozapine dose increases. Intensified CIM monitoring includes daily laboratory assessment and daily ECG. Ideally, a TTE should be performed. The safety threshold is defined as either CRP levels exceeding 100 mg/L or troponin levels exceeding 2× ULN. Both findings are highly suggestive of CIM and necessitate immediate termination of clozapine as well as immediate initiation of a cardioprotective treatment. The observation that CRP increases typically precede troponin elevations by several days [136, 137] implies that the CRP safety threshold might be exceeded first in the majority of cases.

ECG signs of CIM are mostly unspecific and include T-wave inversion and ST elevation or depression [11, 125]. Additionally, ECG might reveal signs of arrythmia arising from myocarditis [144]. TTE might show left ventricular impairments and pericardial effusion [125, 132, 146, 147]. Cardiac magnetic resonance imaging (CMR) is the modality of choice for the non-invasive diagnosis of myocarditis in general and the confirmation of CIM [12, 38, 146,147,148]. Although endomyocardial biopsy (EMB) remains the gold standard for the final diagnosis of myocarditis, this invasive approach cannot be performed in every patient and involves a considerable residual risk for complications. EMB may be required to rule out a viral myocarditis as a differential diagnosis [125, 149]. The CMR-based diagnosis of acute myocardial inflammation relies on multiparametric techniques including T1 and T2 mapping and the presence of late gadolinium enhancement (LGE) [150, 151]. LGE describes a regional injury, whereas T1 and T2 mapping reflect the myocardial changes due to inflammation including myocardial oedema and fibrosis. T1 and T2 mapping allow a robust and quantifiable imaging application for an assessment of the underlying disease [151]. Supporting CMR findings consist of pericardial effusion and inflammation, as well as regional wall motion abnormalities or hypokinesis. Although the underlying mechanism of CIM are not fully understood, the resulting cardiac pathologies including myocardial injury, myocardial oedema and pericardial involvement can be frequently visualized using CMR [26, 148, 152]. To summarize, if available, the use of TTE and CMR (Fig. 2) can aid the diagnosis of CIM and decrease rates of unnecessary discontinuation [10].

Based on the evidence and considerations discussed in this section and mirroring our approach to CIM screening, we propose separate sets of diagnostic procedures in cases of suspected CIM—a minimal set for low-resource conditions and an optimal set for high-resource conditions (Table 2).

4.4 CIM Management

Confirmed CIM necessitates immediate discontinuation of clozapine [10, 12], simultaneous initiation of a cardioprotective pharmacotherapy encompassing angiotensin-converting enzyme (ACE) inhibitors or angiotensin II receptor blockers, beta-blocker [12, 16, 125, 127, 153,154,155] and sodium–glucose cotransporter 2 inhibitors for more efficient prevention of adverse cardiac remodelling [155,156,157]. Additionally, physical exercise must be avoided for at least 3–6 months [38, 154, 158, 159]. Initially, daily laboratory and ECG monitoring are mandatory to detect any further complications. Data regarding the continuation of cardioprotective pharmacotherapy are limited, but existing evidence supports continued treatment to reduce long-term risk of heart failure [160].

Severe forms of CIM might require intensive care treatment [38, 161, 162], pharmacotherapy with digoxin, loop diuretics [153], corticosteroids and intravenous immunoglobulins [161]. Ventricular arrhythmia, persistent heart failure and sudden death constitute severe complications of CIM. They typically result from delayed or missed diagnosis and treatment of CIM [11,12,13, 38]. Conversely, strict adherence to the screening procedures and the laboratory alert and safety thresholds summarized above drastically increases the probability of a timely detection of CIM and prompt cessation of clozapine, which in turn facilitate a full recovery [10, 63, 125].

In cases of inevitable immediate clozapine discontinuation clinicians should be vigilant regarding potential clozapine withdrawal symptoms including cholinergic and serotonergic discontinuation syndrome [163,164,165,166,167]. Peripheral and central manifestations of cholinergic rebound syndrome encompass, among others, diarrhoea, vomiting, delirium, hallucination, dyskinesia and dystonia [163, 164, 168]. Serotonergic symptoms include diaphoresis, hyperreflexia, agitation and clonus [164]. Adequate supportive care, discontinuation of concurrent serotonergic compounds whenever possible and short-term administration of anti-cholinergic compounds and cyproheptadine for serotonergic withdrawal symptoms have been proposed as management strategies [164].

4.5 Communication and Patient Involvement

Appropriate communication of the associated potential benefits, side effects and adverse events as well as patient involvement by means of shared decision-making (SDM) are key for treatment with any antipsychotic [169, 170] but especially for clozapine [7, 171]. This approach has a positive impact on both the likelihood of patients agreeing to a particular form of treatment including clozapine as well as on patients’ treatment adherence [172,173,174]. For CIM detection and management, these issues are especially relevant. Patients will likely be very concerned about their short- and long-term cardiac health, especially considering the extensive diagnostic and therapeutic measures required. By contrast, experienced treatment teams might regard a diagnosis of CIM as a relatively routine situation. However, they need to adequately address patients’ concerns, which might be further exacerbated by their persisting clinical symptoms. Moreover, caregivers should acknowledge the fact that patients’ choices in this situation are severely constrained by the medical necessities associated with CIM. When faced with clozapine discontinuation due to CIM, patients might lose hope, as their best treatment option is suddenly no longer available to them. Here, a decatastrophizing communication style is crucial. Clinicians should emphasize that a full recovery from CIM is very likely and that a re-challenge can be a safe and viable strategy after a minimum waiting period of 3 months [159, 175,176,177,178]. While data on this topic are currently lacking, the way prescribers communicate with and involve the patient in this situation will likely also influence a patient’s willingness to agree to a potential re-challenge.

This also pertains to the way information about the risk for adverse events and their long-term consequences is communicated before clozapine initiation. Prescribers should address these issues already during their initial clozapine offer as openly as possible. Here, the emphasis should be on the high probability to detect adverse events including CIA and CIM early, their good manageability, the low risk of persistent complications and the option of a re-challenge in many cases. Effective and pre-emptive communication on these matters during all stages of clozapine treatment is also essential to counteract the growing impact of inaccurate information patients can obtain from objectively less-credible digital sources including social media [179].

4.6 CIM Prevention

While CIM screening should be an integral part of clozapine treatment, implementing systematic CIM prevention strategies, which are informed by CIM pathophysiology and risk factors, for every initial and repeated clozapine initiation are of even greater importance. These measures should increase patient safety, facilitate outpatient initiation of clozapine and reduce treatment burden as well as permanent clozapine discontinuation. CIM prevention is also crucial because some patients might not consent to a post-CIM re-challenge. CIM prevention might be of greatest significance for lower-income countries, where severely limited resources will likely preclude extensive non-clinical screening procedures. CIM prevention strategies apply to all cases without a prior history of CIM or other types of clozapine-induced inflammation. Specific strategies for post-CIM re-challenges are covered in Sect. 5. While these prevention strategies are crucial for increasing patient safety, we would like to emphasize that they do not constitute a replacement for the CIM screening measures summarized above, such as CRP monitoring.

Almost all known risk factors appear to converge on one central pathophysiological mechanism for CIM, namely rapidly rising initial plasma clozapine levels. Therefore, adjusting clozapine titration regimens and accounting for other factors influencing plasma clozapine levels should be the focal point of CIM prevention efforts. While official manufacturer guidelines in most countries still recommend titration rates of 25 mg per day [21, 45, 133], all available evidence points to the importance of instituting considerably lower titration rates, particularly during the earliest stages of treatment [62, 180].

The paramount importance of slow titration is also underscored by the fact that in countries such as Denmark and the Netherlands, which have widely adopted this strategy, low rates of CIM have been reported [31, 32, 44, 134]. Conversely, data from Australia, where initial daily clozapine dose increases of 25 mg remain the norm, indicate considerably higher CIM rates [11, 25, 30,31,32, 44, 134, 181]. Additionally, it is also essential to minimize the use of co-medication, which reduce clozapine metabolism or otherwise increase CIM risk, whenever possible. Specifically, concurrent treatment with valproate should be avoided, especially during initial clozapine titration. Caution is also warranted with respect to olanzapine [18, 62]. Typically, switches from olanzapine to clozapine requires a plateau cross-taper switch strategy [182]. This entails a period of several weeks, during which patients are exposed to both antipsychotics. In these cases, particularly slow clozapine titration is warranted. Moreover, due to the inhibitory effects on CYP1A2 enzyme activity, strongly elevated CRP levels might require temporarily slowing titration rates [36].

To provide a practical guideline incorporating all relevant aspects, individualized titration regimens accounting for ethnicity, sex , age, obesity, pregnancy, CYP450 enzyme activity status, co-medication (e.g. olanzapine and valproate), smoking behaviour and inflammation have been proposed [62, 78]. Currently, this represents the most sophisticated approach for CIM prevention. Early routine use of therapeutic drug monitoring should further reduce the risk of dangerously fast increases in clozapine plasma levels during initial titration [62]. Consequently, recommendations for community clozapine titration regimens emphasize both slow initial dose titration and intermittent dose plateaus [143].

4.7 Clozapine-Associated Cardiomyopathy

Clozapine-associated cardiomyopathy (CAC) is a highly relevant adverse event in the context of CIM. It is commonly defined as either the new onset of decreased left ventricular ejection fraction (LVEF < 50 %) or a reduction of at least 10% compared with baseline LVEF or the most recent LVEF associated with clozapine treatment. According to a recent meta-analysis, clozapine exposure is associated with a cardiomyopathy event rate of 6 per 1000 individuals and with an absolute death rate from cardiomyopathy of 3 per 10,000 individuals [11]. However, published incidence rates vary considerably, which might be attributable to a lack of standardized diagnostic criteria and procedures [11, 25].

The clinical presentation of CAC most commonly features shortness of breath and palpitations—symptoms generally consistent with heart failure [183]. Yet, a considerable portion of cases may be asymptomatic [147]. CAC symptom onset can vary widely from months to years after treatment initiation [25, 37, 183]. After clozapine discontinuation, which is mandatory, cardiomyopathy improved to a relevant degree in most cases [8, 183,184,185,186,187]. Clinical observations indicate that long-term outcome might depend on the degree of ejection fraction (EF) reduction, with the poorest outcome reported in cases with an EF of < 25 % measured at the time of CAC onset [183]. By contrast, individuals with an EF between 25% and 40% supposedly improve significantly, while EF > 40% can be associated with a quite full recovery [8, 183,184,185,186,187,188]. Cardioprotective treatment encompassing at least an ACE inhibitor and a beta-blocker is an essential element of CAC management [183].

There is pre-clinical evidence for a non-inflammatory mechanism of CAC [8, 60, 147], specifically disturbances of mitochondrial structure and function caused by clozapine’s electrophilic nitrenium ion resulting in decreased oxygen consumption rates [60]. The latter finding suggests a pathophysiological overlap with CIM [56]. However, the pathophysiology of CAC including possible mechanistic similarities with CIM remains poorly understood.

Although a direct link between CIM and the development of cardiomyopathy might be conceivable in cases of undiagnosed myocarditis [25, 37, 76, 189, 190], the scarcity of available data currently precludes any firm conclusions [25, 147]. Myocarditis caused by viral, bacterial, protozoal or fungal infections can lead to inflammatory cardiomyopathy, which may result in left ventricular dysfunction, heart failure or arrhythmia [191]. However, it remains unknown, whether CAC could in some cases be a complication of undetected CIM.

Important general risk factors for cardiomyopathy including diabetes, hypertension, dyslipidaemia, smoking and obesity, which are clearly prevalent in patients requiring or receiving clozapine treatment, always need to be considered as underlying causes [25]. Persistent tachycardia, which can be a consequence of clozapine treatment, constitutes another risk factor [76, 192,193,194,195]. Effective treatment of tachycardia using ivabradine or beta-blockers can both prevent and reverse cardiomyopathy [76, 193,194,195,196]. The latter might also require concurrent use of ACE inhibitors or angiotensin II receptor antagonist. In the context of clozapine treatment this can increase the risk for orthostatic hypotension [76, 129].

5 Clozapine Re-challenges After CIM

There is increasing evidence in favour of re-challenging clozapine after CIM. However, there are currently no prospective studies investigating this issue. Moreover, the procedures of published post-CIM re-challenges vary widely in all important aspects, such as timing, titration speed and monitoring regimens, and include several procedures, which must be regarded as unsafe in the light of current evidence [26, 27, 36, 63, 132, 197,198,199,200,201,202,203]. This underscores the lack of an established consensus and precludes any straightforward synthesis of recommendations founded mainly on published case reports. We therefore base the current set of recommendations primarily on the pre-clinical and clinical evidence reviewed in Sects. 3 and 4, emphasizing maximal patient safety. In our view, this cautious approach is clearly warranted, given that any re-challenge involves patients, which have evinced a particular vulnerability to this potentially lethal serious adverse event. Importantly, the recommended re-challenge procedures require a mental health care setting with sufficient resources. Our recommendations need to be viewed in this light and do not represent a global consensus.

5.1 Why to Re-challenge

The efficacy of the commonly used first- and second-generation antipsychotics is comparable, while side-effect profiles can differ drastically [204]. Therefore, adverse events arising from treatment with a particular antipsychotic would usually result in permanently switching to a different compound. For clozapine, the situation is fundamentally different, because of its unmatched efficacy for several key clinical syndromes, and for reducing hospitalization and mortality rates [5, 204,205,206,207,208,209,210,211,212,213,214,215]. Among all antipsychotics, clozapine elicits the largest improvements in global psychopathology [216,217,218,219], positive [3], negative [216,217,218,219] and depressive symptoms [204, 219].

Compared with other antipsychotics, patients treated with clozapine show greater improvements in treatment adherence [204] and lower rates of suicidal behaviour [220,221,222,223]. Clozapine is also the most effective drug for reducing aggressive behaviour [224,225,226,227,228] and substance use [208, 209, 229]. Perhaps most importantly, clozapine reduces all-cause and suicide mortality [206] to a larger degree than any other antipsychotic [214, 219]. Clozapine is also associated with the lowest risk for extrapyramidal side effects of all antipsychotics [204]. Accordingly, current expert consensus recommendations also specifically include second-line use of clozapine in cases of persistent positive symptoms with co-occurring extrapyramidal symptoms, tardive dyskinesia, suicidality and aggression [133].

This extensive body of evidence underscores the indispensable role of clozapine for the treatment of schizophrenia, which can often make it challenging, if not outright impossible, to find an adequate replacement. The immediate termination of clozapine in confirmed cases of CIM is mandatory [230] and at the very least necessitates a search for a temporary alternative. Both long-acting injectables (LAIs) and electroconvulsive therapy (ECT)—possibly aided by transient use of benzodiazepines—can be considered as preferential treatment options in this case. LAIs have been proposed as alternative treatment options for TRS, especially in the form of high-dose regimes [231]. However, there is no evidence for a level of efficacy approaching clozapine [232, 233]. Although ECT constitutes a safe and effective treatment strategy for clozapine-resistant schizophrenia [233,234,235,236,237,238,239], evidence for sufficient medium- and long-term efficacy of ECT in TRS without concomitant clozapine is lacking. Treatment with other oral antipsychotics after clozapine discontinuation is also clearly not equally effective [240,241,242].

Consequently, even temporarily substituting clozapine with another antipsychotic can have substantial negative clinical and functional consequences [230, 232, 240, 242, 243]. Catatonia and persistent psychotic symptoms appear to be the most common sequelae of abrupt clozapine termination [165, 244, 245]. Moreover, clozapine termination might increase short-term suicide risk [246]. In the long term, both all-cause mortality and the risk for suicide attempts also appear to increase [240, 247].

Typically, terminating clozapine because of CIM results in permanent discontinuation, which is in line with current official recommendations [5, 200]. However, these recommendations do not take into account the potentially severe negative short- and long-term consequences outlined above [10, 232, 240, 247, 248]. Importantly, data from registry studies support the notion that re-initiation of clozapine following all-cause discontinuation is superior to treatment with either oral olanzapine, LAIs or ECT [240, 242].

Based on these findings and the extensive body of evidence reviewed above, a clozapine re-challenge after CIM is increasingly recognized as a crucial alternative strategy. This notion is also supported by the approximately 40 published cases, which indicate an average success rate of approximately 60% [26, 27, 36, 63, 132, 197,198,199,200,201,202,203].

5.2 When to Re-challenge

Post-CIM re-challenges are generally regarded as a viable approach if important prerequisites have been met [26, 27, 36, 63, 132, 197,198,199,200,201,202]. Before a re-challenge can be offered, patients need to have made a full recovery from CIM. Recovery from myocarditis can be assessed reliably after a minimum of 3–6 months [175, 177, 249], a fact that should guide the timing of a re-challenge. Verifying a full recovery requires a diagnostic workup including ECG, TTE and laboratory assessments (CRP, troponin T, NT-proBNP, CK and full blood count) (Fig. 3) [26, 59, 199, 201]. Ideally, CMR should also be performed due to its superior sensitivity for detecting residual inflammation and other subtle cardiac pathologies [12, 26, 38, 146,147,148, 152]. Finally, a cardiological consultation is essential for evaluating patients’ cardiac health. Even if cardiac health was fully restored, avoiding any treatment with a relevant risk for a recurrence of myocarditis will likely be clearly preferable from a cardiological point of view. While these considerations and the diagnostic workup should inform the risk–benefit assessment, the final decision regarding a re-challenge should rest with the patient and the psychiatric treatment team.

Fig. 3
figure 3

Post-CIM re-challenge workflow. Slow titration regimens need to be complemented with clinical, electrophysiological and laboratory screening. Here, the same principles relevant for any clozapine treatment apply. This includes a comprehensive baseline assessment encompassing ECG and TTE as well as CRP, troponin T, NT-proBNP, CK and CK-MB. Frequency, scope and duration of CIM screening measures during the actual re-challenge in published case reports vary considerably. Given that patients’ safety should be paramount, all standard screening measures should be performed at least as often as during a regular clozapine treatment initiation, i.e. at least once per week or even twice. In line with this approach, many published case reports encompass the following: troponin and CRP at least twice per week, vital signs at least once daily and ECG twice per week. If available, this screening should be expanded to TTEs and monitoring of NT-proBNP in cases of suspected CIM. Conducting a CMR during a re-challenge without any clinical or laboratory indication for CIM is not warranted. Furthermore, repeated TDM after every 50 mg of clozapine dose increase without waiting for a steady state is crucial to ensure a sufficiently slow titration regime. This regime should be continued until a therapeutic dose has been reached. BP blood pressure, CIM clozapine-induced myocarditis, CK creatine kinase, CK-MB muscle–brain type CK, CMR cardiac magnetic resonance imaging, CRP C-reactive protein, ECG electrocardiogram, HR heart rate, NT-proBNP N-terminal-pro hormone and brain natriuretic peptide, TDM therapeutic drug monitoring, TTE transthoracic echocardiogram

Full size image

Every post-CIM re-challenge should be based on a SDM process involving fully informed patients, their chosen persons of trust and their potential legal representatives. In many ways, this process should mirror the SDM process, which preceded the initial clozapine treatment, but must also include a detailed risk–benefit assessment. On the benefit side, the course of illness after clozapine discontinuation and possible alternative strategies, which might have previously been effective in a particular patient, need to be taken into account. Importantly, patients need to be aware that, due to the considerable delay of the full therapeutic effects [236, 250, 251], their response to clozapine before discontinuation due to CIM will most likely not reflect the full extent of achievable benefits.

On the risk side, clear measures to ensure patient safety during a re-challenge and to detect a recurrence of CIM in the very early stages constitute crucial prerequisites [59, 202]. Accordingly, post-CIM re-challenges require an appropriate psychiatric hospital setting with direct care of or close supervision by an experienced clozapine team and access to an appropriate cardiological infrastructure and regular consultations [59, 159, 198, 199]. The favourable success rate of post-CIM re-challenges under these circumstances is a key fact that should inform the SDM process.

Conversely, myocarditis as part of DRESS syndrome signifies an increased risk for additional multi-visceral complications [122, 252,253,254] and a considerably lower re-challenge success rate compared with isolated CIM [122, 252]. Therefore, the risk–benefit assessment for a re-challenge after DRESS syndrome with myocarditis cannot be considered favourable at present.

5.3 How to Re-challenge

Of all identified factors, clozapine titration speed has the strongest link to CIM risk. Accordingly, a very slow dose titration regime constitutes the most important strategy for a successful re-challenge [27, 59, 62, 197,198,199, 203]. Given that our knowledge regarding CIM risk factors remains far from complete, consistently using a very slow titration speed appears to be most prudent. Therefore, we propose an initial clozapine titration rate of 6.25 mg every 4 days. Recommendations for slow regular outpatient clozapine titration feature titration rate increases after a certain daily dose has been reached [143]. Mirroring this approach, we propose clozapine titration rate increases to a rate of 6.25 mg every 3 days after reaching a daily dose of 75 mg and to a rate of 6.25 mg every 2 days after reaching a daily dose of 150 mg (Fig. 3). These stepped titration rate increases should reduce the overall time required for a re-challenge without compromising patient safety.

This titration rate should minimize the impact of concomitant medication associated with an elevated CIM risk such as valproate and olanzapine. Avoiding these compounds whenever possible should increase re-challenge success rates [25, 27, 36, 199]. However, in some cases this might not be feasible, for instance during a plateau cross-taper switch from olanzapine to clozapine [36, 62, 182]. Vaccinations with an inherent increased risk for myocarditis such as SARS-CoV-2 should not be performed in the weeks before or during a re-challenge [96,97,98,99,100,101,102, 255].

Slow titration regimens need to be complemented with extensive clinical, electrophysiological and laboratory screening [59, 63, 159, 199, 201]. This includes a comprehensive baseline assessment encompassing ECG and TTE as well as CRP, troponin T and NT-proBNP [59, 63, 159, 199, 201]. Because patient safety should be paramount, all standard screening measures should be performed at least as frequently as during a regular clozapine treatment initiation, i.e. at least once per week or even twice.

In line with this approach, many published case reports include the following screening measures: troponin and CRP at least twice per week, vital signs at least once daily and ECG twice per week [27, 59, 132, 159, 197,198,199, 201, 203, 256, 257]. This screening should be expanded to TTEs and monitoring of NT-proBNP in cases of suspected CIM [27, 59, 159, 199, 256]. Conducting a CMR during a re-challenge without any clinical or laboratory indication for CIM is not warranted. Furthermore, repeated TDM after every 50 mg of clozapine dose increase without waiting for a steady state is crucial to verify a sufficiently slow titration speed.

For regular clozapine treatment, the high-risk period for CIM of approximately 8 weeks provides the rationale for the standard CIM screening duration [15, 59, 199, 256]. Conversely, post-CIM re-challenges are likely associated with a considerably longer CIM high-risk period. First, all patients involved have been shown to be at particularly high risk for CIM. Second, the slow dosing titration regime—while likely protective for most patients—might delay manifestation of CIM in some cases. Consequently, screening duration should be extended until a therapeutic dose is reached. Ideally, at this stage a follow-up CMR can help to rule out any recurrence of cardiac pathologies. Figure 3 provides a comprehensive summary and timeline of all advisable procedures under optimal conditions.

Currently, there are no established procedures for the long-term follow-up after post-CIM re-challenges, beyond the standard long-term monitoring mandatory for all clozapine patients [5, 27, 59, 201]. Any signs of cardiac dysfunction should prompt a new diagnostic workup for CIM and cardiomyopathy.

6 A Global Perspective

Unfortunately, clozapine utilization remains a global challenge [6]. A related issue is the considerable heterogeneity regarding crucial aspects of its actual use including prerequisites for treatment initiation, titration regimes, monitoring guidelines, management and reporting of severe adverse events [29, 133, 134]. Biological, environmental and economic factors relevant for clozapine treatment, which might vary across countries, cannot fully explain this phenomenon. Rather the current state appears to reflect an unrealized potential for harmonizing clinical practices, which results from widely differing opinions regarding the relevance of serious cardiac adverse events related to clozapine and the importance of clozapine titration speed [258].

This is particularly relevant for preventing, detecting, and managing CIM. Currently, practices for titrating clozapine differ substantially across countries [65, 133, 134]. For instance, the Netherlands and Denmark favour slow titration in outpatient settings as do the Maudsley TREAT recommendations [31,32,33,34, 143]. For other European countries, no comparable strategies have been implemented. As outlined above, guidelines in Japan are very restrictive and require clozapine initiation in an inpatient setting [65, 259]. Based on the slow clozapine metabolism in people with Asian ancestry [18, 65, 66], a maximum titration rate of 25 mg per week for Japanese patients has been proposed [65, 260]. It should be noted, however, that considerably slower titration regimes will likely be necessary to better protect a population with a clearly elevated CIM risk. This also applies to the manufacturer-approved titration rates in Australia [30], which remain widely used and may be in part responsible for the high rates of reported CIM [258]. These examples illustrate that the crucial issue of clozapine dosing speed has so far clearly not received the global recognition it deserves [258], even though addressing it should significantly reduce the burden of adverse events without a substantial impact on hospital length of stay [30].

CIM screening practices during clozapine initiation also vary across countries [133, 134]. In the UK, troponin assessments are recommended but not regarded as a mandatory screening procedure [134]. By contrast, the Maudsley TREAT recommendations include an extensive CIM screening [143]. Guidelines in Australia, New Zealand, Canada and Germany deem weekly CRP and troponin assessments to be mandatory during the first 4–6 weeks [134, 261,262,263,264]. Notably, despite the restrictive registration practices and stringent monitoring protocols, in Japan troponin and CRP screening is not mandatory [65, 134, 259]. Guidelines in Denmark and The Netherlands also do not regard either CRP or troponin screening as mandatory [134, 265]. Rather, both laboratory tests are only recommended as part of a diagnostic workup in cases of suspected CIM.

Maybe most importantly, there are stark differences in availability of resources for people with schizophrenia across the world. The scarcity of mental health resources in developing countries can make clozapine treatment particularly challenging, exacerbating the problem of clozapine underutilization already observed in industrialized countries [6, 260, 266,267,268]. This highlights the importance of CIM prevention efforts for low-income countries, where CIM screening and post-CIM re-challenge protocols suitable for industrialized countries will not be feasible. These constraints should not further limit clozapine use.

Hence, a central question is whether CIM screening is strictly necessary, if all established measures for preventing CIM have been implemented. In such a scenario, CIM screening might increase treatment burden unnecessarily. However, to our knowledge there are no studies comparing CIM incidence rates in low-risk settings with and without CIM screening. Some of the studies reporting low CIM rates in conjunction with slow titration rates—including those from the Netherlands and Denmark—were conducted in settings without routine CIM laboratory screening, which might lead to an underdiagnosis of CIM [29, 32]. Data from Turkey demonstrate that introducing systematic CIM screening procedures is associated with increased reporting rates of CIM [269].

Moreover, the long-term consequences of relatively benign, undetected forms of CIM remain unknown. Until these issues are clarified, eliminating CIM screening might be considered as too risky. Notably, despite its outpatient setting and the use of a slow titration scheme the procedures advocated by the Maudsley TREAT team feature an extensive CIM screening including CRP, troponin, BNP and TTE [143]. This implies that patient safety should take precedence, if sufficient resources are available.

However, in low-income countries trade-offs between screening recommendations and broad availability of clozapine treatment may be inevitable. Considering the ubiquitous requirements of weekly blood monitoring for CIA after clozapine initiation, the treatment burden for patients is not increased by the addition of blood-based CIM screening. It has been suggested that CRP might be suitable as the sole screening parameter because its increase typically precedes troponin elevations by several days [136, 137], thus obviating the use of troponin screening [62, 270]. Assessing troponin would only be required in cases of suspected CIM to verify an emerging cardiac pathology [137]. Such an approach would minimize the economic burden of CIM screening.

The use of CMR is another issue illustrating the economic and infrastructural constraints on adhering to gold-standard procedures. CMR is clearly the best non-invasive method for detecting functional and morphological cardiac irregularities as well as myocardial tissue pathology [271]. However, its limited availability and relatively high cost severely restrict its use. Most likely for that reason, even the majority of published post-CIM re-challenges did not utilize CMR.

7 Areas for Future Research

Increasing clozapine utilization and optimizing treatment and monitoring algorithms remain important topics for clinical research. In the context of CIM prevention, screening, detection and management, several topics emerge as particularly relevant. These unresolved issues might explain why no expert consensus regarding CIM-screening and re-challenges has been reached [133].

Progress is likely impeded by clozapine’s status as a generic drug, which results in a lack of interest in and financial support for prospective studies from the pharmaceutical industry. Furthermore, harmonization of clinical practices for clozapine treatment remains a challenge, primarily relying on consensus building driven by the accumulating evidence of investigator-initiated studies. This underscores the importance of establishing national clozapine networks [7]. They appear to be best suited not only to facilitate the implementation of the current state of the art in national guidelines and but also to effect widespread changes in clinical practice. Ideally, this should be accomplished in close collaboration with the TRRIP working group [133, 272], which has emerged as the most influential forum for international consensus building regarding TRS and clozapine use [133].

Systematic implementation and evaluation of CIM prevention measures have the potential to reduce the initial treatment burden associated with CIM screening while also reducing clozapine discontinuation. Here, personalized CIM risk-adjusted titration regimes constitute a valuable basis, which can be refined as further evidence regarding CIM risk factors accrues [62]. However, the slower dose titration, which is central to any CIM prevention strategy, might also discourage patients and prescribers eagerly expecting faster clinical improvements. Demonstrating that reducing initial titration speed improves patient safety will be key to promote acceptance of this approach. Data from countries regularly employing slow clozapine titration in outpatient settings suggest that this strategy lowers myocarditis risk of clozapine to a level comparable with other antipsychotics [31, 32, 44, 134]. However, more data are necessary to draw firm conclusions given that these findings might also partly be attributable to a lack of systematic CIM screening in some countries.

This is particularly pertinent because the impact of clinically inapparent forms of adverse events caused by clozapine-induced inflammation including CIM, which would most likely be overlooked without appropriate screening measures, on long-term health remains largely unknown. Moreover, shifting the focus to the full spectrum of clozapine-induced inflammation while acknowledging CIM as its most prominent manifestation rather than an isolated phenomenon while emphasizing preventability of these adverse events might help to increase awareness regarding the relevance of these issues. This should include establishing consensus standards for the diagnosis and report of adverse events associated with clozapine-induced inflammation. It might also promote systematic screening efforts even in countries with low reported rates of these adverse events. These measures might further enhance the considerable positive impact of clozapine on all-cause mortality.

Moreover, conclusively establishing the sensitivity and reliability of the most relevant screening parameters, i.e. CRP and troponin [62, 136, 137, 270] should help to optimize the balance between patient safety and the burden of CIM screening. Investigating these issues might help building a consensus regarding the minimal requirements for cardiac monitoring. In conjunction with establishing clozapine titration speeds which minimize the risk for inflammatory adverse events, these efforts should help to increase the safe use of clozapine, especially in low-income countries.

Additionally, identifying genetic and additional non-genetic CIM risk factors will be crucial to improve CIM prevention and to facilitate more targeted screening algorithms. This also applies to elucidating the risk mechanisms underlying concurrent treatment with olanzapine, quetiapine and valproate. Given the high rates of treatment resistance in childhood- and adolescent-onset schizophrenia, more data on CIM prevalence and risk factors in this patient group are clearly needed [273,274,275].

To our knowledge, only a limited number of reports have been published of re-challenges after probable CAC, which remitted after clozapine discontinuation [8, 183,184,185,186,187, 276]. Presently, due to the very limited number of cases, re-challenges after clozapine-associated cardiomyopathy are not recommended [8, 133]. This underscores the urgent need for research on CAC and its potential links to CIM.

8 Conclusion

To summarize, there is clear, converging evidence for the singular importance of slow dosing of clozapine for cardiac safety during the initial treatment with clozapine and during post CIM re-challenges. The latter constitutes an important and safe procedure, if accompanied by an appropriately intensified myocarditis screening. CIM prevention efforts adhering to this principle should have the biggest positive impact on patient safety and clozapine utilization. Combining this strategy with establishing a safe minimum CIM screening regime should increase its use globally, including in health-care systems with scarce resources. Another crucial issue pertains to minimizing the negative impact of clozapine discontinuation after CIM. Here, establishing CMR as a standard clinical tool for the detection of CIM and for monitoring its sequelae should facilitate the swift and safe re-initiation of clozapine. Finally, the impressive extent of research on the cardiac effects of clozapine underscores the importance of the issues discussed in this review for addressing clozapine underutilization. These efforts are a testament to the singular status of this decades-old antipsychotic, which remains indispensable for many patients.