A 12-month audit of clozapine associated myocarditis in a South Australian Local Health Network: The importance of screening and personalised titration

Clozapine is effective in up to 50 % of patients resistant to other antipsychotics. Its use is restricted to third-line due to adverse effects which include myocarditis. Australia reports the highest incidence of clozapine-associated myocarditis (CAM) in the context of pharmacovigilance and relatively rapid titration. An audit of patients commenced on clozapine within the Central Adelaide Local Health Network (CALHN) between 2012 and 2015 found an incidence of 8.6 %. We present here a case series from a follow up audit considering titration relevant risk factors for CAM. We reviewed anecdotal cases and data from all hospital-based commencements of clozapine across CALHN for the period July 2021 to June 2022 using pharmacy and medical record databases. We identified 5 cases of CAM and all had risk factors impacting on clozapine metabolism, including rapid titration, elevated baseline CRP, Asian ethnicity and concomitant treatment with inhibitors of clozapine metabolism. While personalisation of clozapine treatment needs further investigation in prospective trials, slower titration to lower targets for risk groups may not impact on hospital length of stay and has the potential to significantly reduce the burden of adverse events. Australian manufacturer approved titration rates exceed those recommended for personalised dosing and may not be safe for patients with risk factors. Early clozapine levels at week two could identify slow metabolisers for dose adjustment. Closer ties between psychiatrists and cardiologists are critical for the development of protocols for safely maintaining clozapine treatment during low level cardiac inflammation and to support safe rechallenge.


Introduction
Clozapine is an atypical antipsychotic with unique efficacy that is restricted to third line use in patients with treatment resistance due to a range of potentially life-threatening side effects that require close monitoring (Clark et al., 2014).While the major focus of monitoring is to prevent agranulocytosis, there are a range of more common complications that carry higher risk of mortality including pneumonia and clozapine associated myocarditis (CAM) (Cohen et al., 2012;de Leon, 2023).CAM is an inflammation of the myocardium that occurs most commonly in the first months of clozapine treatment and is variably associated with tachycardia, symptoms of influenza-like illness and fever, elevation of inflammatory markers for example C-reactive protein (CRP) and markers of cardiac damage such as troponin in peripheral blood.Finally, echocardiography shows a reduction in ejection fraction, which can be associated symptoms of heart failure that is reversible in days with clozapine cessation (Ronaldson et al., 2012a).
Internationally rates are of CAM low at <0.07 % -0.7 %, particularly in continental Europe, however in Australian incidence exceeds 8 % in some studies but the disease is less severe with lower mortality (Dawson et al., 2018;De Las Cuevas et al., 2022;Rohde et al., 2018).High rates in Australia have been attributed to widespread awareness and pharmacovigilance programs (Dawson et al., 2018;de Leon et al., 2022b).Reliance on highly sensitive blood-based markers in these screening programs raises the potential for false positive diagnoses (Clark et al., 2023).However, more rapid titration, promoting inflammation, cytokine release and inhibition of CYP1A2 clozapine metabolism, and higher plasma levels may also contribute to incidence of CAM (de Leon et al., 2023;de Leon et al., 2020b).While IgE-mediated hypersensitivity, similar to Stevens Johnson's syndrome which occurs after rapid lamotrigine initiation, is one hypothesised mechanism (de Leon, 2022), the dose dependent nature of CAM and reports of safe rechallenge (Richardson et al., 2021), suggest other pathways.Other proposed mechanisms include: Cytokine release (TNF-alpha), oxidative stress associated with elevated catecholamines, direct toxicity similar to anthracycline induced cardiomyopathy, comorbid exposure to illicit drugs or alcohol, cholinergic receptor dysfunction and genetic variations in clozapine metabolism (Patel et al., 2019;Vickers et al., 2022).In Australia titrations more rapid than the official regimens, and coprescription of sodium valproate have been associated with increased risk of myocarditis (Ronaldson et al., 2012b).Locally in South Australia (SA), we have identified a CAM incidence of 8.6 % over the first 3 months of treatment in the Central Adelaide Local Health Network (CALHN) (Nachmani Major et al., 2020).We found variability in assessment and management.Such high rates of CAM are concerning and significantly reduce the opportunity for gold standard clozapine treatment of resistant psychosis, reducing the quality of care.
Our SA clozapine monitoring protocol supports local manufacturer recommendations for weekly monitoring for CAM including physical observation for fever, tachycardia, changes in blood pressure and increases in C-reactive protein (CRP) and troponin from baseline to week 4 of titration.Urgent troponin, CRP, ECG and clinical review is indicated if body temperature reaches 38 • C (eCPMS, 2019; SA Health, 2022).Recommended clozapine starting dose is 12.5 mg, titrating to 200 mg daily by day 14 (eCPMS, 2019;SA Health, 2022).This rate is 1/3 faster than some European (outpatient) and Japanese protocols (Netherlands Clozapine Collaboration Group, 2013;Kikuchi et al., 2023).Some ethnic groups, particularly those of East Asian origin (ranging from Pakistan to Japan (de Leon et al., 2020a)) have concentration/dose (C/D) ratios that suggest poor (slow) metabolism of clozapine and appear to be particularly vulnerable to adverse events associated with higher blood levels (de Leon, 2023).de Leon et al. have proposed that optimal titration for patients of Asian descent in the presence of other metabolic inhibitors may rbe limited to 75 mg by day 14 (de Leon et al., 2022b).Recent evidence suggests even slower titration rates were associated with a significant reduction in a range of inflammation related adverse events in a Japanese multisite sample (adjusted OR 4.01; 95 % CI 2.02-7.97;p < 0.001) (Kikuchi et al., 2023).Cases meeting criteria for Drug Reaction with Eosinophilia and Systemic Symptoms (DRESS) syndrome, including hyper-eosinophilia, rash, and evidence of organ damage were exclusively identified in the group with more rapid titration that that suggested by de Leon et al.
To better personalise clozapine initiation and reduce rates of adverse events, de Leon et al. (de Leon et al., 2022b) have recently proposed a guideline for personalised titration designed to optimise plasma levels based on ethnic risk of slow metabolism, interactions with drugs known to increase levels via inhibition of CYP enzymes including sodium valproate, quetiapine, smoking cessation, and factors that indicate a proinflammatory state such as raised CRP and obesity.The titration rates proposed in this guideline are significantly slower than current SA protocols.Here we describe CAM cases identified in a repeat 1 year audit of new clozapine commencements in CALHN, focussing on the adherence to current SA Health clozapine prescribing guidelines and presence of modifiable risk for myocarditis.

Methods
The audit was prompted by concerns over a cluster of 3 anecdotal reports of CAM within CALHN, which encompasses a band of the Adelaide metropolitan area, running east-west, including the central business district and is home to around 481,000 people or 1/3 of Adelaide's population.Mental Health services include 2 main public hospitals, and stand-alone mental health facility, as well as community-based outpatient services.These services care for around 400 patients treated with Clozapine.The study was approved as quality improvement project by both SA Pharmacy and the CALHN Mental Health Quality and Governance Committee and subsequently as an audit by the CALHN Human Research Ethics Committee (approval reference 18201).
Electronic and paper records were accessed to review investigations and standardised clinical documentation for anecdotal cases.Subsequently, a systematic audit of electronic records of all patients treated in CALHN hospitals over this period using data extracted from iPhar-macy®; OACIS® and Sunrise ® databases, augmented by paper-based record review where required, from July 2021 to June 2022 was performed to identify additional cases of CAM.Fig. 1 outlines the search process for myocarditis cases.For all clozapine treated patients, duplicate entries and patients admitted to health services outside of CALHN were removed and blood test results for troponin and CRP testing performed during the study period were reviewed.To capture evolving cases we reviewed in more detail any participants with troponin above SA Pathology thresholds of 12 ng/L in females and 16 ng/L in males.All participants blood testing was performed at SA Pathology laboratories.Only patients with a formal diagnosis of CAM in the discharge summary were included in the final analysis.We collected clinical and investigation data based on Ronaldson et al. criteria (Ronaldson et al., 2011), represented in the SA Health clozapine management guideline (SA Health, 2022).Recommendations included daily clinical and blood monitoring for patients that develop mild troponin increase from 1 to 2 times normal, CRP between 50 and 100 mg/L or HR over 120 bpm (Ronaldson et al., 2011).Clozapine should be ceased when troponin exceeds 2 times the upper limit of normal or CRP of >100 mg/L.CRP was analysed using high sensitivity assays with a normal reference range < 8 mg/L.Clozapine metaboliser status was assessed based on the presence of modifying factors such as obesity, inflammation, interacting drugs and ethnicity, all which may modify cytochrome p450 activity (de Leon et al., 2022b).Clozapine level/dose (C/D) ratios were calculated based on the average of the previous 5 days dosing, where clozapine levels were available (Ertugrul et al., 2022;Koenig et al., 2022).

Results
Medical record search identified 217 individual patients hospitalised on clozapine treatment during the study period, 53 with elevated Troponin T results greater than 12 ng/L in females and 16 ng/L in males, 35 of these were admitted to CALHN inpatient units, 14 with accompanied elevation in CRP.Four cases received a discharge diagnosis of CAM.From anecdotal reports only one out of 3 cases received a discharge diagnosis of CAM, for a total of 5.
For the 5 cases of CAM identified, Table 1 shows key characteristics and risk factors.Cases were 60 % male, age range 19-44 years.Abnormal troponin levels were first identified on day 16-20.Cardiac MRI was used to confirm diagnosis in 3 cases, echocardiography in 2, both in one.The was no confirmatory cardiac imaging documented for one case (case 4) and in another (case 5) echocardiography was negative but CAM still diagnosed.All cases included cessation of clozapine as part of management.All cases had at least 1 risk factor for elevated clozapine levels.One case was of Asian ethnicity, 2 cases had elevated CRP on initiation, 2 cases were titrated at above local guideline recommended rates, 1 patient was a smoker but smoking is prohibited in SA inpatient services, 80 % were obese and the remaining case overweight, and all were treated with concomitant medication that could potentially interact with clozapine to increase risk of myocarditis.Personalised titration rates according to de Leon et al. (de Leon et al., 2022a) were exceeded in 4/5 cases.Maximum recommended personalised dose was exceeded in 2 cases (2 and 5).Supplementary tables 1-5, summarise dose, physical and testing parameters at the onset of CAM.

Case 1
19-year-old Caucasian male, obese (BMI 30.72), smoker (6/day), past use of illicit substances and alcohol dependence.Failed previous treatment with both olanzapine and paliperidone.At baseline the patient had an incidental finding of reduced left ventricular ejection fraction (LVEF) (47 %) noted during screening prior to clozapine commencement.Managed with cardiology support and the commencement of bisoprolol and ramipril prior to titration.Baseline troponin 4 ng/L, CRP 1.3 mg/L, ECG normal.Clozapine was titrated according to SA Health protocol, cross titrated from olanzapine, however initial regular observations were difficult due to agitation.Sodium Valproate commenced day 10 due to mood instability/irritability with features of intermittent explosive disorder.Patient was admitted to a closed unit when sodium valproate was initiated, and hence smoking cessation was enforced.First symptoms of myocarditis on day 14 when the patient developed a febrile illness with coryzal symptoms.Clozapine was ceased on day 18 following raised troponin.Cardiac MRI on day 30 showed a LVEF of 49 %.Diagnosis was myocarditis and pericarditis due to clozapine, differential viral peri-myocarditis.Cardiology suggested rechallenge was possible if LVEF remained >40 %.Supplementary table 1 shows the evolution of critical parameters.Risk factors in this case included, obesity, smoking cessation, and the commencement of valproate.No clozapine levels were taken during titration.This case was subsequently successfully rechallenged without valproate (switched to lamotrigine and lithium) at a slightly slower titration rate (75 mg by day 8, 200 mg by day 18), after 18 months of ineffective management on paliperidone LAI plus oral olanzapine.Troponin remained below threshold across re-titration with mild CRP elevation from day 7, peaking at 29.2 by day 14 and normalizing by day 21.

Case 2
44-year-old Caucasian, female, obese (BMI 37.8), ex-smoker, alcohol dependency (14-22 drinks/day), history of untreated psoriasis including face.Previous failed treatment with amisulpride, asenapine, lurasidone, aripiprazole, and olanzapine.Baseline troponin 3 ng/L, CRP 24.3 mg/L, ECG and Echocardiography normal.On clozapine commencement taking quetiapine, paliperidone depot and paroxetine.Commenced on standard clozapine titration after alcohol detoxification.Quetiapine titrated and ceased by day 8. Shifted to more rapid titration from day 9 with target of 250 mg by day 14.Discharged to community but represented on Day 20 to the emergency department.Supplementary table 2 shows that temperature, CRP, troponin and clozapine levels were substantially elevated at representation.The patient was noted to have a Staphylococcus Aureus bacteraemia on blood culture (day 19) likely secondary to osteomyelitis from a previous left tibial fracture internal fixation, culture negative on subsequent repeat, but treated with IV flucloxacillin.Repeat echocardiography was normal but cardiac MRI showed a small region of subepicardial late gadolinium enhancement of the basal anterolateral left ventricular wall with small amount of adjacent pericardial fluid, suggestive of myocarditis, without evidence of infarction.While a diagnosis of CAM was made and clozapine ceased, this blood culture result suggests an alternative diagnosis of S. aureus associated myocarditis.Potentially this case could be rechallenged.
Clozapine levels were very high, peaking at 1630 μg/L.Risk factors for CAM included bacteraemia and elevated baseline CRP, obesity, and treatment with quetiapine and paroxetine.

Case 3
27-year-old Caucasian male, overweight (BMI 26.1), recent diagnosis of schizophrenia, complicated by low mood and suicidality.Fluoxetine 20 mg commenced on admission.Symptomatic hyperprolactinemia secondary to risperidone hence treated with additional aripiprazole.Risperidone was down titrated and ceased aripiprazole increased to 20 mg mane.Clozapine was titrated according to standard protocol.Baseline troponin 3 ng/L, CRP 0.4 mg/L, ECG and echocardiography normal.Supplementary table 3 indicates that the patient experienced chest pain on day 18 when elevated troponin was first noted and clozapine was ceased on day 19.Cardiac MRI day 22 was consistent with myocarditis.Risk factors at baseline included treatment with fluoxetine.No clozapine levels were taken during titration.

Case 4
44-year-old Caucasian male, obese (BMI 30.5), recent pulmonary embolism, cholestatic liver function abnormalities, treated with paliperidone depot and the anticoagulant apixaban at clozapine titration.History of schizophrenia, cerebral palsy, major depression, posttraumatic stress disorder, and cluster B traits.Incomplete response to risperidone, paliperidone, zuclopenthixol, amisulpride, aripiprazole, and olanzapine depot.Baseline troponin 3 ng/L, CRP 11.7 mg/L, ECG and echocardiography normal.No potential source or further investigation of this mildly elevated baseline CRP was documented.The patient was titrated as per standard protocol.Supplementary table 4 shows CRP elevated on day 13, Troponin T on day 16.Clozapine ceased with diagnosis of asymptomatic CAM.Cardiology suggested rechallenge could be attempted with Beta Blocker prophylaxis, however, clozapine has not been recommenced.Risk factors for CAM included obesity, mildly elevated baseline CRP, abnormal liver function.No clozapine level was taken during titration.

Case 5
37-year-old female of Asian background, obese (BMI 30.4), exsmoker, past use of illicit substances.History of failed treatment with risperidone (ceased due to extrapyramidal side effects), lurasidone (mania), and olanzapine.At baseline troponin <3 ng/L, CRP 6 mg/L.During titration the patient was treated with olanzapine 20 mg and the oral contraceptive pill (Levlen).Supplementary table 5 shows that Echocardiography was normal day 14.Titration was more rapid than guidelines reaching 250 mg by day 15.Patient was discharged and reviewed on day 18 by primary care physician and amoxicillin/clavulanic acid 875/125 mg BD was prescribed for respiratory tract infection.Clozapine levels were elevated on day 19 above consensus maximum level of 1000 μg/L (SA Health, 2022).Patient presented to an emergency department on the evening of day 20 febrile with a 2-week history of cough.Echocardiography day 21 appeared normal, although the cardiology opinion was not favourable.The patient was successfully rechallenged with clozapine plus paliperidone LAI after 6 months unsuccessful treatment with oral olanzapine.Limited detail was available but after 6 weeks of treatment the patient was stable on 300 mg/ day with a clozapine/ norclozapine level of 410/192.Risk factors for CAM included Asian ethnicity, obesity, infection, cotreatment with olanzapine and oestrogen containing OCP (Levlen which combines oestrogen (ethinylestradiol) and progesterone (levonorgestrel)) and rapid titration.No clozapine level was taken during titration.

Discussion
CAM cases identified in our audit displayed a range of modifiable risk factors including baseline inflammation and drug interactions that potentially combined to inhibit metabolism and generate high plasma levels during clozapine titration.These factors all occured in the context of relatively rapid Australian titration rates, which although supported by manufacturer recommended local guidelines, are faster than international sites with much lower rates of CAM. de Leon et al. (de Leon et al., 2022a) suggest that those with obesity, taking oestrogen containing oral contraceptives (OCP), sodium valproate, flupentixol, olanzapine, perphenazine or quetiapine, or those of Asian or native American descent, are suspected poor metabolizers and should be titrated slowly.Obesity alone was present in 4/5 of our cases and potential medication interactions were present in 3/5 including olanzapine, sodium valproate, quetiapine, paroxetine and oestrogen containing OCP. Olanzapine is primarily metabolised at CYP 1A2 hence has the potential to inhibit clozapine metabolism (de Leon et al., 2022a).Olanzapine has independent risk for myocarditis (De Las Cuevas et al., 2022).The interaction between clozapine and sodium valproate metabolism is complex, however, valproate acts as an inhibitor during clozapine titration (de Leon et al., 2022a).Both sodium valproate and quetiapine have been identified as risk factors for myocarditis in Australian cohorts (Nachmani Major et al., 2020;Ronaldson et al., 2012b).Quetiapine has also been associated with the severity of myocarditis (De Las Cuevas et al., 2022).Paroxetine, is an inhibitor of CYP2D6 that can increase clozapine levels by 30 % (Diaz et al., 2008;Spina et al., 2016;Wijesinghe, 2016).Oestrogen containing OCPs are known to inhibit CYP1A2 and CYP2C19 (minor clozapine metabolism) and elevate levels (Gabbay et al., 2002;Schoretsanitis et al., 2020).
Based on de Leon et al. (de Leon et al., 2022a), Table 1 shows that proposed personalised target doses are much lower for the 4/5 poor metabolisers identified, particularly for females and non-smokers of Asian origin, maximum dose may be as low as 75 mg allowing for inflammation risk and medication interactions.In SA clozapine is most often commenced in hospital hence there may be some concerns that slow titration rates would lengthen inpatient stay.However, slower titration is coupled with lower target dose that is suggested to be achieved within 3-4 weeks, similar to current Australian guidelines.Personalised dose targets are based on pharmacokinetic principles to predict plasma drug levels close to the gold standard of 350 μg/L for efficacy (de Leon et al., 2022a).Clozapine levels as low as 250 μg/L may be effective in managing treatment resistant psychosis (Northwood et al., 2023) and anecdotally some patients can be well managed at lower plasma levels.One recent Japanese study suggests slower rates and low doses are effective in Asian populations (Kikuchi et al., 2023).However, more evidence is required to prospectively validate the effectiveness of these recommendations.Potentially doses may be safely increased beyond low targets slowly once clozapine induced inflammation reduces and tolerance develops.
Interestingly, cotreatment with paliperidone LAI, occurred in 3/5 cases.There is emerging evidence for use of Paliperidone LAI and clozapine in clozapine resistant cases (Bioque et al., 2020).Paliperidone is largely renally cleared with some minor metabolism at CYP 2D6 and 3A4 overlapping with minor clozapine metabolism (Wijesinghe, 2016).These patients all had other risk factors for slow metabolism that likely contributed to negative outcomes.
The single case without prominent risk factors, case 3, did receive cotreatment with medications known to inhibit minor pathways for clozapine metabolism: Aripiprazole (CYP 3A4) and fluoxetine which is considered a clinically relevant inhibitor with a complex profile: CYP2D6 (strong), CYP2C9 (moderate), CYP2C19 and CYP3A4 (weak to moderate) and CYP1A2 (weak effects on clozapine metabolism) (Spina and de Leon, 2014;Wijesinghe, 2016).The additive effects of multiple weak CYP inhibitors may result in clinically significant changes in clozapine levels (Ruan et al., 2020).As further evidence of the range of interactions associated with slow clozapine metabolism accumulates, this combination may come to be considered problematic.Although no clozapine levels were taken to suggest metabolic rate, this case may have been a genetic poor metaboliser of clozapine.Such mutations have not been well explored, however CYP1A2*6 and CYP1A2*7 have been associated with poor metabolism in European samples (Ruan and de Leon, 2020;Shelton et al., 2022).The identification and genotyping of such individuals is critical to improve prediction of those at risk of adverse events.
There were gaps in guideline adherence for inflammation screening, titration rates and clozapine level monitoring.We have previously identified that investigation and management of patients with risk factors and features of CAM is variable and requires close collaboration between cardiology and psychiatry (Nachmani Major et al., 2020).Of concern 2/5 cases displayed CRP elevations at baseline, suggesting gaps in knowledge of the risks of clozapine titration in the context of preexisting inflammation.Particularly case 2 where titration was continued in the presence of a bacteraemia.There appeared to be no investigation of the elevated CRP in case 4. Clozapine itself causes inflammation that inhibits its own metabolism at CYP1A2, hence additional inflammatory burden at titration onset is likely to increase the risk of early high levels and adverse outcomes (Clark et al., 2018).Potentially in two cases, early discharge was associated with less monitoring, in the context of risk factors for CAM, suggesting that higher risk patients may need closer follow up.The SA clozapine guidelines currently suggest clozapine levels at week 4 or prior to discharge.Adherence to this direction was incomplete but where taken levels were high.Given emergent risk in these cases appears from week 2, clozapine level monitoring should commence at this time or earlier to assist in identifying slow metabolisers with high levels at risk of adverse events.In the cases identified here, the use of cardiac imaging was not standardised and while clozapine was ceased for all patients, two were subsequently successfully rechallenged after a further period of poor response to other agents.There is evolving evidence for rechallenge after CAM or other clozapine-related inflammatory events, addressing risk factors, at slower titration rates and with prophylaxis from cardioprotective medications such as beta-blockers (Griffin et al., 2021;Shivakumar et al., 2020).Given the opportunity cost of clozapine cessation better protocols for the safe management of CAM are urgently required (Clark et al., 2023).

Limitations
This audit was retrospective and suffers with associated biases and missing data on some variables that may have been used in decision-making but could not be located on review.Analysis of electronic data was triangulated across systems and confirmed with paper-based records where possible.Gaps in data occurred when patients were treated at sites not linked to the main hospital electronic health record, including one hospital and the community sites.Community based mental health services in Adelaide also use private pathology providers not linked to the system.In these cases data was dependent on paperbased records.All hospital sites are recently linked and further prospective study is required using the SA State-wide monitoring system (Clark et al., 2014).The scope of this audit did not allow a full review of cases with troponin elevation but no formal CAM diagnosis.Potentially, these cases may result from subclinical inflammation that could have been avoided with slower titration (Clark et al., 2023;de Leon, 2022).

Conclusions
The majority of CAM identified in this study showed modifiable risk factors that could have been considered prior to clozapine initiation to reduce the risk of inflammation, high levels and adverse events.Close attention to exisitng screening guidelines is also critical to minimise adverse events.Earlier clozapine levels at week two may assist in identifying slow metabolisers in a timely fashion for dose adjustment.Australian manufacturer approved clozapine titration rates exceed those recommended for personalised clozapine dosing and may not be safe for patients with risk factors.While personalisation of clozapine titration needs further investigation in well-designed prospective trials, careful consideration of ethnic diversity in metabolism, inflammation promoting factors and drug interactions is compelling.Slower titration to lower targets for risk groups may not impact on hospital length of stay and has the potential to significantly reduce cost and burden of adverse events during clozapine titration.The efficacy of clozapine treatment at lower dosing in these patients needs more exploration.In 2 out of these 5 cases, patients were successfully re-titrated at slower rates.Potentially the remaining 3 cases could also be rechallenged.Further collaboration between psychiatrists and cardiologists is required to develop better guidance for optimal cardiovascular outcomes of clozapine titration, including the maintenance of clozapine treatment during mild titration related inflammatory responses, and to support safe rechallenge.

Contributors
OC conceptualisation, data collection, drafting of manuscript; SRC conceptualisation, drafting and review of manuscript; LRW conceptualisation, review of manuscript; KN conceptualisation, supervision and review of the manuscript; VT conceptualisation, supervision and review of the manuscript.

Role of funding source
This work was unfunded.

Declaration of competing interest
SRC has participated in advisory and educational boards and received speaker's fees from Janssen-Cilag, Lundbeck, Otsuka, and Servier; research funding from Janssen-Cilag, Lundbeck, Otsuka, and Gilead; and data sharing from Viatris Australia.All other authors report no conflicts of interest.

Table 1
Demographics and risk factors for CAM cases, C-reactive protein -CRP, long-acting injection -LAI, Echocardiography -Echo, Oral Contraceptive Pill -OCP.
Leon et al., 2022a;de Leon et al., 2022benged.bBased on De leon et al., female non-smokers are thought best managed at the lower end of the personalised dose range (deLeon et al., 2022a;de Leon et al., 2022b).cSAHealth guideline recommendation is 200 mg at 14 days.From day 14 the dose can be increased in 50 mg intervals every two to three days depending on efficacy and side effects.Maximum dose is 900 mg per day.O.Carswell et al.