Clinical course and analysis of ten fatal cases of clozapine-induced myocarditis and comparison with 66 surviving cases

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h re s Clinical course and analysis of ten fatal cases of clozapine-induced myocarditis andcomparison with 66 surviving cases Kathlyn J. Ronaldson , Paul B. Fitzgerald , Andrew J. Taylor , Duncan J. Topliss John J. McNeil a Department of Epidemiology and Preventive Medicine, Monash University, The Alfred Centre, 99 Commercial Road, Melbourne, VIC 3004, Australiab Monash Alfred Psychiatry Research Centre, Alfred Hospital and Monash University, School of Psychology and Psychiatry, P.O. Box 315, Prahran, Melbourne VIC 3181, Australiac The Heart Centre, Alfred Hospital, P.O. Box 315, Prahran, Melbourne VIC 3181, Australiad Department of Endocrinology and Diabetes, Alfred Hospital, P.O. Box 315, Prahran, Melbourne VIC 3181, Australia Background: Fatal clozapine-induced myocarditis has not been investigated systematically. We describe the clinical course of 10 fatal cases of myocarditis with clozapine and identify factors associated with fatality.
Methods: Cases of myocarditis were documented from the patient's medical records and fatal cases also from Results: The fatal cases of myocarditis occurred 1996–2009 and were diagnosed at autopsy. Before death,three had no symptoms of illness and only three had cardiac-specific diagnostic results. None was investigated by cardiac imaging techniques, and in none was myocarditis suspected before death. Duration of clozapine for the fatal cases was 14–33 days with an outlier at 4.5 months. Only 3 cases had significant Comparison of these ten cases with 66 non-fatal cases indicated no significant difference in gender, age, smoking status, dose at onset or concomitant sodium valproate. However, obesity (BMI N 30 kg/m2) was significantly more frequent among fatal than non-fatal cases (60% vs 26%; p b 0.03) and duration of clozapinewas significantly longer for fatal cases (20.8 vs 17.0 days; p b 0.006), after exclusion of one outlier. Creatinekinase (CK) N 1000 U/L was also associated with death (p = 0.0004).
Conclusions: Routine monitoring for myocarditis for the first 4 weeks of clozapine, and discontinuation ofclozapine in the presence of evidence consistent with myocarditis may assist to prevent fatalities occurringfrom early-onset myocarditis. Investigation by cardiac imaging will give a measure of severity and need forintervention. Obesity may increase the risk of mortality and CK N 1000 U/L may indicate life-threateningillness.
2011 Elsevier B.V. All rights reserved.
creased dramatically. Consequently, when reviewedthe cases reported to the Australian Therapeutic Goods Administra- The association between myocarditis and therapeutic use of tion up to the end of 2003, using numbers of patients registered with clozapine has been known with some confidence at least since the the two sponsor companies, they estimated an incidence of 0.7–1.2%.
regulatory agencies in the United Kingdom ( Another Australian estimate of about 2% was based on data from a in a review article have avoided estimating the incidence of ) published alerts on the subject. A subsequent article by clozapine-induced myocarditis, citing a number of reasons for the risk , describing 15 cases of clozapine-related being substantially underestimated, including low rates of reporting myocarditis occurring in Australia in the first month of therapy, to spontaneous adverse reaction reporting databases, variability in the provided supportive evidence for the association, with the incidence presentation of myocarditis making diagnosis difficult, and the low of fatality from this cause a thousand-fold increase on the background likelihood of psychiatric patients seeking medical attention for physical illness. The data presented by suggest Since the publication of the paper by the the mortality rate with cases of myocarditis is about 10%.
awareness of clozapine-induced myocarditis in Australia has in- reported five fatal cases and subsequently individual fatal cases have been described by . Fatal cases have also been listed in seriesbased on data collected by a national adverse reaction reporting ⁎ Corresponding author. Tel.: +613 9903 0840; fax: +613 9903 0556.
0920-9964/$ – see front matter 2011 Elsevier B.V. All rights reserved.
doi: K.J. Ronaldson et al. / Schizophrenia Research 128 (2011) 161–165 have occurred in the United States up to 2001 ( Characteristics of the 10 fatal and 66 non-fatal cases of clozapine-induced myocarditis.
In addition, described a fatal overdose in which myocarditis was diagnosed at autopsy.
To date there has been no review specifically of fatal cases. In this paper we describe the clinical course of ten fatal cases, compare key characteristics with those of surviving cases, and seek to identify factors which may serve as clues to the prevention of myocarditis- related fatality in patients starting clozapine.
Fatal and non-fatal cases of clozapine-related myocarditis were a Data on smoking status available for 9 fatal and 61 non-fatal cases and on BMI for all identified from reports submitted to the Therapeutic Goods Admin- istration, the Australian drug regulatory authority, between January1993 and December 2009; details of other cases were communicated to the authors by staff of the health services for which ethicscommittee approval had been obtained (see 2.1); and fatal cases The time from commencement of clozapine to death was 14–34 alone were identified from cases in the National Coroners' Information (mean 21.4) days with an outlier at 136 days (4.5 months). One patient had discontinued clozapine at the time of death (Case 1).
Documentation of each case included the date of birth, sex, Another had missed a dose 3 days before death, and then had a clozapine start and end dates, dose of clozapine taken each day, reduction in dose from 100 to 50 mg/day (Case 7). The others were baseline pathology results and results obtained during the course of continuing clozapine without interruption.
clozapine treatment and until death or resolution of symptoms for Six patients were showing clinical signs of illness at the time of surviving cases, signs and symptoms recorded in the progress notes and death. Two of the cases, perhaps three, were apparently well, including details described in the autopsy report. Documentation for each case Case 3 who reported that he was “feeling great” on the day before he was reviewed by the study steering group for compliance with the case died overnight in his sleep. Cases 4, 5 and 6 were thought to have upper definition for clozapine-related myocarditis ).
respiratory tract infections, Case 7 was investigated for neurolepticmalignant syndrome and Case 1 had mild fever and renal impairment (creatinine 250 μmol/L). Case 10, which occurred after 4.5 months,had been admitted non-specifically unwell both physically and Approval for the study was obtained from the Human Research mentally 10 days before death. It is unclear whether this illness was Ethics Committees of the following institutions and health services: related to myocarditis, but the post mortem indicated acute and Monash University, Austin Health, Barwon Health, Bayside Health, Bendigo Health, Department of Human Services, Department of Justice, Only four patients had cardiac-specific investigations but Eastern Health, Mercy Health, North West Mental Health, Peninsula none had echocardiography or other direct investigation of cardiac Health, St Vincent's Health, Southern Health (all from Victoria); Northern Sydney Central Coast Health, Sydney South West Area Health Case 9 is notable for having died of myocarditis following re- Service (Royal Prince Alfred Hospital and Concord Hospital Zones) initiation of clozapine, having used clozapine previously without any (from New South Wales); and Prince Charles Hospital (from Queens- indication of an adverse effect, although documentation was insuffi- land). The approvals covered access to medical records without patient cient to confidently exclude a missed episode of myocarditis during consent. In addition, an Access Agreement was signed with the Victorian Institute of Forensic Medicine for access to the National Coroners' The plasma clozapine concentrations (0.1–2.7 mg/L) among the Information System database and a Deed of Confidentiality and Conflict fatal cases reflect the prescribed clozapine dose of each individual. In of Interest with the Therapeutic Goods Administration for access to no case was alcohol or any illicit drug found in the post mortem blood.
A key feature of these cases is that myocarditis was not suspected in any of these patients prior to death.
Statistical analysis was conducted using STATA/IC version 10. Risk ratios as for a cohort study were calculated for categorical variables, with Autopsy indicated that five of the fatal cases were free of coronary p-values based on the Chi-squared test. The 2-tailed Students t-test was artery disease, two had minimal atheroma and three had significant used to investigate the difference in means for continuous variables.
disease with stenosis greater than or equal to 50% Thosewith BMI greater than 30 kg/m2 were equally divided between those with significant atherosclerosis and those free of disease. The autopsyreport for Case 4, who had the most severe atherosclerotic narrowing, Myocarditis was diagnosed at autopsy in each of the 10 fatal cases recorded that ischaemic heart disease and hypertension had contrib- by the presence of mixed inflammatory infiltrates in cardiac histology.
uted to the death as independent causes. Case 9 was found to have All except one case had hyperinflated and/or congested lungs, and thromboemboli occluding both pulmonary arteries and myocarditis was a secondary cause. In no other case was pre-existing cardiovas- The age and sex distribution of the fatal cases is indicated in cular disease considered to have contributed to the deaths.
presents further characteristics of the same cases andthe clinical and diagnostic details. The ten cases occurred 3.3. Comparison of fatal and non-fatal cases between 1996 and 2009. Six patients had a body mass index (BMI) ofmore than 30 kg/m2 and seven (of nine; data missing for one) were compares characteristics of the fatal cases and the non- fatal cases. There was no significant difference between fatal and non- K.J. Ronaldson et al. / Schizophrenia Research 128 (2011) 161–165 Table 2Characteristics of the 10 fatal cases of myocarditis.
fatal cases for gender, age, proportion of smokers, dose at onset or proportion taking sodium valproate. However, the risk ratio for deathwith BMI above 30 kg/m2 was 3.5 (95% CI 1.1–10.9; p b0.03) and Compared with surviving cases, there is nothing remarkable about duration of clozapine was significantly longer (p b0.006) for the fatal the fatal cases in age, sex, smoking status, last clozapine daily dose and cases compared with those that were non-fatal, even after removal of concomitant medication. Sodium valproate was included in the analysis because it was the only medication other than clozapine In addition, all three of the fatal cases for which results were taken by more than two fatal cases. The apparent association of a BMI available had very high creatine kinase (CK) results (N1000 U/L). In of more than 30 kg/m2 with a greater likelihood of death may be a contrast, only 3 of 65 surviving cases had a CK value of more than random occurrence, but it is biologically plausible that cardiac 1000 U/L (p = 0.0004; Fisher's exact test), and only 12 had results compromise in the setting of obesity reduces the prospect of recovery.
exceeding twice the upper limit of normal.
The significantly longer mean duration of clozapine use for fatal cases Table 3Clinical and diagnostic features of the 10 fatal cases of myocarditis.
Case Pro-dromal Heart rate Other clinical observations NMS, PE and serotonin 50% stenosis of left anterior Abbreviations: bpm, beats per minute; CK, creatine kinase; CK MB, creatine kinase muscle and brain; CRP, C-reactive protein; FBE, full blood examination; NA, not available; NMS,neuroleptic malignant syndrome; PE, pulmonary embolism; RR respiration rate; WBC, white blood cells; ULN, upper limit of normal.
K.J. Ronaldson et al. / Schizophrenia Research 128 (2011) 161–165 may be an indication that fatality followed delayed recognition of published case, clozapine was withdrawn when the 64-year-old illness and delayed withdrawal of clozapine.
woman developed tachycardia, fever and hypoxia She nevertheless died 4 days later of sudden cardiorespiratory reported a range of plasma levels for patients currently taking arrest. Fulminant myocarditis was diagnosed at autopsy. Likewise one clozapine of 0.0–2.61 (mean 0.34; 114 patients) mg/L. In a study by of the patients among the present ten had stopped taking clozapine post mortem clozapine blood concentrations following seven cases of fatal self-poisoning were 3.7–12 mg/L in Since the fatal case occurring after 4.5 months of clozapine is the only individuals who had been poorly compliant or who had used case among 76 with onset later than clozapine day 33, late developing clozapine prescribed for another person. These literature comparisons cases appear to be very rare. Hence, it is difficult to conceive of a cost- suggest that there is no reason to believe that clozapine toxicity was a effective monitoring scheme for a death occurring so late after clozapine factor in the ten fatalities caused by myocarditis.
commencement. However, it is possible that the chronic myocardial The high CK results for three cases are notable, in comparison with inflammation found at autopsy originated during the first 4 weeks of data for the non-fatal cases and in view of the absence of data for the clozapine and the monitoring employed was inadequate to effect other fatal cases. found that higher CK values (mean for 11 cases 1708 ± 1217 U/L) were associated with fulminant myocarditiscompared with milder non-fulminant disease.
4.1. Coronary artery disease and fatal myocarditis In the month following the publication of the article by Kilian et al.
in November 1999, Novartis Australia circulated a cardiac monitoring Since the study was retrospective, it was not possible for us to protocol in Australia. These guidelines have been followed in varying reliably document pre-existing cardiovascular disease and risk factors degrees across Australia since that date. Six of the ten cases described for such disease, such as hypertension, diabetes and hyperlipidaemia, here occurred in the context of use of these guidelines and consequent to compare these characteristics in fatal and non-fatal cases. However, awareness among clinicians and pathologists of the possibility of we were able to ascertain that only 50% of the fatal cases had some death from myocarditis in patients recently initiated on clozapine.
degree of atherosclerosis, and for only one of these was ischaemic Hence, there may be an enhanced likelihood of fatal myocarditis in heart disease considered to have contributed to the death. persons taking clozapine at the time of death being identified at reviewed a cohort of 134 individuals with schizophrenia who autopsy in Australia, compared with other countries.
died aged 48.6 ± 14.8 years and whose families had donated brain recently published an observational study tissue to the Maryland Brain Collection in the US state of Maryland.
comparing rates of sudden cardiac death occurring in the community Sixty eight percent had died of natural causes, including 45.7% who in patients taking antipsychotic drugs among Medicaid enrollees in died of cardiovascular disease. The frequency of atherosclerosis among the US state of Tennessee. Although the exposure for clozapine was these individuals, determined at autopsy, was 44%, and is similar to low, the incident rate ratio was higher for clozapine than for other that observed among these cases of fatal clozapine-induced myocar- antipsychotics. It is feasible that at least some of these deaths may ditis. It is apparent that coronary artery disease was not a major factor have been caused by myocarditis. An early study by in the death from myocarditis of these cases.
using mortality data for 196 current users of clozapine acrossthe US did not identify myocarditis as a cause of death, but conduction disorders or sudden death and respiratory illness occurred at twiceand three times the rate, respectively, in current versus past users.
The question arises, would any of these ten deaths have been Five deaths were attributed to pneumonia in an analysis conducted in preventable if more had been understood about clozapine-induced the United Kingdom by of causes of death for 21 myocarditis at the time of presentation? Since myocarditis was not patients taking clozapine. In this study, cause of death was obtained suspected in any of these cases, it may be that more care with from case notes or death certificate and the number of deaths monitoring for this outcome during the first 4 weeks after com- investigated by autopsy was not specified. Since patients developing mencement of clozapine and/or taking appropriate steps in response myocarditis frequently have the symptoms of a respiratory tract to clinical and diagnostic indicators that myocarditis may have been infection, autopsy findings typically reveal congested and/or hyper- present would have prevented some of these deaths. The longer inflated lungs, and myocarditis may be missed even if the myocar- duration of clozapine exposure for the nine fatal cases occurring in the first 5 weeks of clozapine compared with the non-fatal would support is possible that some of these fatalities with a diagnosis of pneumonia the view that stopping clozapine earlier may have prevented death.
or respiratory illness may have been missed myocarditis.
Even those who do not develop obvious clinical signs and symptoms of illness may be identified if routine monitoring is followed. Although prevention of serious illness and fatality with monitoring is unlikely tobe as reliable as for symptomatic cases, three of the non-fatal cases were Despite the risk of death from myocarditis, a recent study asymptomatic and for these myocarditis was detected using routine including all patients with a diagnosis of schizophrenia in Finland monitoring. The case who had previously taken clozapine without any found that clozapine was associated with a significantly lower all- indication of the development of myocarditis illustrates the importance cause mortality than any other antipsychotic medication of monitoring reintroduction of clozapine with just as much care as In other studies, protection against mortality had been There remains, however, the possibility that in these patients there but an implication from the Finnish study is that this may be an were factors such as severity of the inflammatory assault on the incomplete explanation. Further, in a meta-analysis, clozapine rated myocardium and host factors governing the response of the heart to above other second generation antipsychotics on two out of three this assault that may have meant that the patient would have measures of efficacy plus overall symptoms and was second only to succumbed despite a prompt response to rises in cardiac enzymes or amisulpride and sertindole for quality of life This left ventricular impairment measured by echocardiography. In the evidence of the superior efficacy and safety of clozapine is an incentive absence of documentation of such parameters and/or suitable to investigate clozapine-induced myocarditis further so that it can be response to these results, this possibility remains speculative. In one prevented, or at least the consequences minimised.
K.J. Ronaldson et al. / Schizophrenia Research 128 (2011) 161–165 journal, Internet: . Accessed26 Nov 2010.
Committee on Safety of Medicines, 1993. Myocarditis with antipsychotics: recent cases This is the first study to examine the clinical features associated with clozapine (Clozaril). Curr. Probl. Pharmacovigilance 19 (3), 9–10 Internet: with fatal clozapine-related myocarditis, and to compare them with those of surviving cases. The nine early fatal cases illustrate the Conley, R.R., Shim, J.-C., Kelly, D.L., Feldman, S., Yu, Y., McMahon, R.P., 2005.
necessity of routine monitoring for myocarditis in patients commenc- Cardiovascular disease in relation to weight in deceased persons with schizophre- ing clozapine. Monitoring should not be conditional on the presence nia. Compr. Psychiatry 46 (6), 460–467.
of clinical symptoms, but it should be intensified in those who develop Feldman, A.M., McNamara, D., 2000. Myocarditis. N. Engl. J. Med. 343 (19), 1388–1398.
Fineschi, V., Neri, M., Riezzo, I., Turillazzi, E., 2004. Sudden cardiac death due to a physical illness within 4 weeks of commencing clozapine. Those hypersensitivity myocarditis during clozapine treatment. Int. J. Leg. Med. 118 (5), with laboratory results suggestive of myocarditis should have clozapine discontinued, to prevent further injury. Cardiac imaging Flanagan, R.J., Spencer, E.P., Morgan, P.E., Barnes, T.R.E., Dunk, L., 2005. Suspected clozapine poisoning in the UK/Eire, 1992–2003. Forensic Sci. Int. 155 (2–3), 91–99.
techniques may assist to assess the seriousness of impairment of Haas, S.J., Hill, R., Krum, H., Liew, D., Tonkin, A., Demos, L., et al., 2007. Clozapine- systolic function and the need for drug therapy for cardiac failure or associated myocarditis: A review of 116 cases of suspected myocarditis associated supportive measures. High creatine kinase results (N 1000 U/L) may with the use of clozapine in Australia during 1993–2003. Drug Saf. 30 (1), 47–57.
Hagg, S., Spigset, O., Bate, A., Soderstrom, T.G., 2001. Myocarditis related to clozapine be indicative of a high risk of mortality, and obesity may impair the treatment. J. Clin. Psychopharmacol. 21 (4), 382–388.
heart's capacity to recover from myocarditis, but these factors warrant Hill, G.R., Harrison-Woolrych, M., 2008. Clozapine and myocarditis: a case series from further investigation. Death from clozapine-induced myocarditis is the New Zealand Intensive Medicines Monitoring Programme. NZ Med. J. 121(1281), 68–74.
independent of coronary artery disease, except in cases where this is Kilian, J.G., Kerr, K., Lawrence, C., Celermajer, D.S., 1999. Myocarditis and cardiomy- opathy associated with clozapine. Lancet 354 (9193), 1841–1845.
La Grenade, L., Graham, D., Trontell, A., 2001. Myocarditis and cardiomyopathy associated with clozapine use in the United States. N. Engl. J. Med. 345 (3), A salary was paid to Dr Ronaldson in 2006 and 2007 from a grant from the Lee, C.H., Tsai, W.C., Hsu, C.H., Liu, P.Y., Lin, L.J., Chen, J.H., 2006. Predictive factors of a Australian National Health and Medical Research Council (NHMRC). Professor fulminant course in acute myocarditis. Int. J. Cardiol. 109 (1), 142–145.
Fitzgerald is supported by an NHMRC Practitioner Fellowship. The NHMRC had no Leucht, S., Corves, C., Arbter, D., Engel, R.R., Li, C., Davis, J.M., 2009. Second-generation input into study design, interpretation of data or decision to publish.
versus first-generation antipsychotic drugs for schizophrenia: a meta-analysis.
Lancet 373 (9657), 31–41.
Medsafe Editorial Team, 1995. Clozapine and myocarditis. Prescriber Update (9) 7–8.
Referenced in: Medsafe Editorial Team, 2000. Potentially fatal complications of Professor McNeil conceived of the study of clozapine and myocarditis and has had a clozapine therapy: myocarditis, venous thromboembolism. Prescriber Update 20, supervisory role in the conduct of the project. Dr Ronaldson conducted the data collection and analysis and wrote the paper. Dr Taylor was the adjudicator determining whether each potential case met the case definition. Professor Fitzgerald proposed Meeker, J.E., Herrmann, P.W., Som, C.W., Reynolds, P.C., 1992. Clozapine tissue analysis that could be conducted, using the data collected. All authors contributed to concentrations following an apparent suicidal overdose of Clozaril. J. Anal. Toxicol.
and approved the final shape of the manuscript.
Meltzer, H.Y., Alphs, L., Green, A.I., Altamura, A.C., Anand, R., Bertoldi, A., et al., 2003.
Clozapine treatment for suicidality in schizophrenia: International Suicide Prevention Trial (InterSePT). Arch. Gen. Psychiatry 60 (1), 82–91.
Professor McNeil has acted as a consultant for Mayne Pharma, one of the sponsors Merrill, D.B., Dec, G.W., Goff, D.C., 2005. Adverse cardiac effects associated with for clozapine in Australia, and for Hospira Australia after the takeover in 2007. Dr clozapine. J. Clin. Psychopharmacol. 25 (1), 32–41.
Ronaldson, Prof Fitzgerald, Dr Taylor and Prof Topliss have declared no conflicts of Ray, W.A., Chung, C.P., Murray, K.T., Hall, K., Stein, M.C., 2009. Atypical antipsychotic drugs and the risk of sudden cardiac death. N. Engl. J. Med. 360 (3), 225–235.
Ronaldson, K.J., Taylor, A.J., Fitzgerald, P.B., Topliss, D.J., Elsik, M., McNeil, J.J., 2010.
Diagnostic characteristics of clozapine induced myocarditis identified by an analysis of 38 cases and 47 controls. J. Clin. Psychol. 71 (8), 976–981.
The authors are grateful for the assistance of Professor J Duflou, Chief Forensic Taylor, D.M., Douglas-Hall, P., Olofinjana, B., Whiskey, E., Thomas, A., 2009. Reasons for Pathologist, Department of Forensic Medicine, NSW Health & Clinical Professor, discontinuing clozapine: matched, case–control comparison with risperidone long- University of Sydney for identifying some of the deaths occurring in New South Wales acting injection. Br. J. Psychiatry 194 (2), 165–167.
Tiihonen, J., Lonnqvist, J., Wahlbeck, K., Klaukka, T., Niskanen, L., Tanskanen, A., 2009.
11-year follow-up of mortality in patients with schizophrenia: a population-basedcohort study (FIN11 study). Lancet 374 (9690), 620–627.
Tirupati, S., 2006. Clozapine and heart in the Hunter region. Aust. NZ J. Psychiatry 40 (1), Annamraju, S., Sheitman, B., Saik, S., Stephenson, A., 2007. Early recognition of Walker, A.M., Lanza, L.L., Arellano, F., Rothman, K.J., 1997. Mortality in current and clozapine-induced myocarditis. J. Clin. Psychopharmacol. 27 (5), 479–483.
former users of clozapine. Epidemiology 8 (6), 671–677.
Australian Adverse Drug Reactions Advisory Committee, 1994. Clozapine and myocarditis. Aust. Adverse Drug React. Bull. 13 (4) Listed in the index for the

Source: http://ccretherapeutics.org.au/research/clozapine/schizresearch_article.pdf