Adolescent with Treatment-Refractory Schizophrenia and Clozapine-Induced Cardiomyopathy Managed with High-Dose Olanzapine

Chief Complaint and Presenting Problem

L . is a 17-year-old bilingual Chinese adolescent male with a history of disorganized schizophrenia, four prior psychiatric hospitalizations, multiple failed medication trials, and a history of clozapine-induced cardiomyopathy (CIC). He was transferred from an acute-care psychiatric hospital to a long-term state children's psychiatric hospital for further treatment of visual and auditory hallucinations, violent and aggressive outbursts, bizarre delusions, and thought disorder.

History of Present Illness

It is unclear when L.'s psychiatric symptoms first presented. According to his father, L. had no psychiatric symptoms prior to age 13 years when he immigrated to the United States from China. During his first year in the United States, L. showed no signs of mental illness but struggled with poor school performance. By age 14, L.'s family began to notice gradual changes in his mood. Initially, L. appeared depressed, and his family attributed this to poor school performance, missing his mother (who remained in China), and other challenges of acculturation. By age 15, he continued to perform poorly in school, started to appear internally pre-occupied, and reported auditory hallucinations. He was referred to a child and adolescent psychiatrist and initially treated as an outpatient with risperidone 6 mg/day for 11 months. However, L.'s symptoms persisted, and he was switched to olanzapine 10 mg/day for 11 months. At that time, L. further decompensated and wandered away from school, resulting in his first inpatient psychiatric hospitalization at age 16.

At intake, L. reported experiencing auditory hallucinations since age 7 years. During the first hospitalization, L. was initially continued on olanzapine 10 mg/day. Lorazepam 1.5 mg/day was added for catatonia because L. was holding a bottle in the air for an extended period of time. L. also appeared paranoid and confessed to thoughts of killing his father in order to save himself. L. was subsequently cross-tapered from olanzapine back to risperidone 6 mg/day. During the hospitalization, L. showed minor improvements in agitation, paranoia, and internal pre-occupation. An electroencephalogram (EEG) did not show epileptiform activity, ruling out seizures as the source of psychosis. L. was continued on risperidone and discharged to outpatient care at a bilingual (Chinese/English) clinic where he was treated for several months.

As an outpatient, L. demonstrated persistent psychotic symptoms while continuing to take risperidone 6 mg/day. L. began to further decompensate, reporting auditory hallucinations and stating he heard “Lots of voices of American men … they speak to me in the train station, telling me where to sit, and they speak to me at night …” He also engaged in bizarre behavior such as excessively washing his groin, walking around with wet pants, covering his genitalia in public, and frequently looking behind himself. L. was hospitalized for a second time due to this decompensation three months after discharge at the age of 16 years.

During L.'s second psychiatric hospitalization, in addition to auditory hallucinations and thought disorder (which ranged from tangentiality to word salad), L. endorsed the delusion that he was a cartoon character and that a “historical puzzle” hit him when he was lying in bed. Initially, risperidone 6 mg/day was continued and amantadine 100 mg twice daily (BID) was added for akathisia. However, risperidone was soon discontinued as it was found ineffective. Risperidone was cross-tapered with ziprasidone over the next two weeks, and amantadine was continued. An electrocardiogram (ECG) was performed prior to starting ziprasidone and indicated potential left ventricular hypertrophy, but a follow-up echocardiogram showed normal biventricular size and function, with mild aortic regurgitation. However, despite treatment with ziprasidone for a total of 5 weeks (two weeks at 80 mg BID) and amantadine 100 mg BID, L. remained actively psychotic. A decision was made to start clozapine, since L. remained very symptomatic despite three atypical antipsychotic medication trials; however, olanzapine had not been increased beyond 10 mg/day, and ziprasidone treatment was limited to two weeks at 80 mg/day.

During this second hospitalization, L. showed minor improvements on clozapine; however, he continued to pace and laugh to himself in the hallway and manifested substantial lack of energy, social withdrawal, and paucity of thought. The clozapine dose was maintained at 350 mg/day, as higher doses caused severe sedation. Amantadine for akathisia was discontinued. Aripiprazole 10 mg/day was subsequently added to clozapine after 19 weeks of clozapine monotherapy, in an effort to improve his prominent negative symptoms. Six months after admission, L. was deemed “adequately organized” despite symptoms of isolation, social withdrawal, concrete thinking, and auditory hallucinations. The inpatient team discharged L. to an adolescent day-treatment program.

While attending the day-treatment program, L. appeared disorganized and self-absorbed. He endorsed command auditory hallucinations with a voice that instructed him to grab faces off of dead people, and transfer their faces to his face. L. was continued on clozapine 350 mg/day and aripiprazole 10 mg/day. During four months in the day-treatment program, L. developed sialorrhea, and his sedation worsened. He often overslept and was frequently late for school. As a result, the dose of clozapine was lowered to 275 mg/day. However, L. subsequently developed disruptive behaviors, drew odd cartoons, and he continued to pace. Aripiprazole was discontinued after nine weeks, and benztropine 0.5 mg/day was added for akathisia.

L. remained in the day-treatment program for four months when he was admitted for his third inpatient psychiatric hospitalization at age 17 years. He endorsed auditory hallucinations of a male voice telling him to kill himself and hurt others. L. was disruptive and irritable; he reportedly threw things at the wall, broke his handheld video game, banged his head, and continued to pace. On admission, L.'s clozapine dose was 275 mg/day and the clozapine plasma level was 241 ng/mL. Clozapine was increased by 25 mg/day for 4 days to 375 mg/day. During the titration, benztropine was discontinued and clozapine was augmented with haloperidol 10 mg/day and clonazepam 0.5 mg/day for ongoing symptoms of psychosis and agitation.

On the 8th day of hospitalization, L. appeared heavily sedated; he complained of subjective fever, and his blood pressure decreased from about 110/80 to 90/60 mmHg. Notable was a tachycardia of 120–130 beats per minute, his ECG showed a mildly prolonged QTc interval, creatine phosphokinase (CPK) was elevated at 462 U/L [normal: 45–245 U/L] and an ECG showed reduced left ventricular function with left ventricular hypertrophy. The shortening fraction was 25% [normal: 30%–42%]. L. was diagnosed with CIC and was quickly tapered off clozapine. Of note, troponin I and a complete blood count (CBC) (including eosinophil count) were within normal limits. Clozapine was replaced with haloperidol 7.5 mg BID and clonazepam 0.75 mg/day. The inpatient team reported that haloperidol was partially effective in reducing L.'s agitation and perseverative behavior, and L. no longer endorsed hallucinations. He was considered adequately stabilized on haloperidol 7.5 mg BID and clonazepam 0.75 mg/day. After approximately one month, L. was discharged to a day-treatment program.

According to his family, L. was compliant with his medications following discharge. However, by four days after discharge, L. had again started to decompensate, and he soon became illogical and irritable. Six days later, he was hospitalized for the fourth time in an inpatient psychiatric hospital for uncontrollable violence, including breaking the family's television and computer and assaulting his father, as well as increased disorganization and illogical speech. During L.'s fourth hospitalization, he was continued on haloperidol 15 mg/day and clonazepam 1 mg/daily, with minimal improvement. Benztropine 1 mg BID was added for neck stiffness and sialorrhea.

L. was discharged after six weeks and subsequently transferred to a children's state psychiatric hospital. On admission, L.'s Brief Psychiatric Rating Scale (BPRS) score was 82.

Past Psychiatric History

There was no past psychiatric history other than described in the present illness.

Developmental History

L. was the product of an uncomplicated pregnancy. He weighed approximately 3.5 kg at birth, and he met developmental milestones on time.

Educational History

L. started school at age 4 years, with some separation anxiety. He attended a rural school in China until age 13 years. According to his father, L. performed in the top five of his class. His father reported that L. was a very good student until he immigrated to the United States, at which time he was diagnosed with a learning disability and “emotional disturbance.” L. attended public bilingual special education classes until age 16 years.

At age 16 years, L. received neuropsychological testing on the inpatient psychiatric unit. L.'s Chinese language skills were assessed at the fifth grade level. His non-verbal intelligence quotient (IQ) was mildly impaired, and his verbal IQ was in the borderline range.

Social History

L. was born and raised in a rural Chinese village. He lived with his parents and maternal grandparents. At age 13 years, L. immigrated to the United States with his father. Mother remained in China as she awaits a visa. Father reported that L. was generally a quiet child. L. had a few close friends while in China but has not made any lasting friendships in the United States. L. has never used illicit drugs or alcohol. L. is fluent in Cantonese and moderately proficient in English.

Family History

There is no reported family history of psychiatric illness on either pedigree.

Medical History

L. was hospitalized in the first month of life for a high fever, which doctors reported to be an allergic reaction to infant formula. He quickly recovered without evidence of sequelae. L. developed hyperprolactinemia while on risperidone. He has a history of mild aortic regurgitation. He also has a history of cardiomyopathy while taking clozapine (shortening fraction 25%), which resolved. L.'s most recent ECG, during his fourth hospitalization, showed shortening fraction 33%–34% [normal: 30%–42%].

Medications on Admission

L. is currently taking haloperidol 5 mg three times a day (TID), benztropine 1 mg BID, clonazepam 0.25 mg at 10 am and 6 pm, and clonazepam 0.5 mg at bedtime.

Mental Status Exam on Admission to State Hospital

L. is a thin Asian male who appeared younger than stated age. He appeared visibly anxious, paranoid, and internally preoccupied. L. presented with marked psychomotor agitation and restlessness. Intermittently, he paced around the room, and at one point he stood up, walked near the door and began to squat repeatedly. L. eventually went back to his seat. He later stood up and punched his father in the face without apparent cause. L. was uncooperative with the mental status exam and generally did not answer questions, even with the help of a Cantonese language translator. He did not endorse or deny hallucinations, paranoia, nor suicidal or homicidal ideation. L. was alert, but appeared internally preoccupied. At one point, L. was asked a question and he repeated it verbatim. According to the translator, much of what L. said was illogical, and he demonstrated disorganized speech in both Chinese and English. L. was unable to cooperate with tests of attention and memory. L.'s insight and judgment were both impaired. His physical exam was unremarkable, and there was no evidence of gait abnormality or motor deficits. There was no evidence of extrapyramidal effects.

Hospital Course

On admission, L. presented with akathisia, agitation, disorganized thought process, visual and auditory hallucinations, and violent and aggressive behavior. At intake, L. struck his father's face and demonstrated catatonic posturing and echolalia. Admission medications included haloperidol 5 mg TID, benztropine 1 mg BID, and clonazepam 1 mg/day. Within the first 14 days of the hospitalization, L. assaulted two peers. L. explained his first assault by stating that he believed that he and his peer were video game characters. However, after he kicked the other patient, he said he realized that she was real and not a cartoon. L. also engaged in bizarre behavior such as repeated "martial arts posturing," engaging in fights with internal stimuli, and repeatedly brushing his teeth in the water fountain. In addition, L. described a delusional system involving Chinese gangsters that were looking for him.

At eight weeks after admission and four weeks after restarting olanzapine, L. reported that the visual hallucinations had disappeared, and his thought process appeared improved. Haloperidol, clonazepam and benztropine were tapered during hospital days 14–73, 28–49 and 28–70, respectively, and replaced with olanzapine and propranolol (for akathisia).

On admission, L.'s BPRS score was 82 on haloperidol 15 mg/day. On day 52 of admission and day 27 on olanzapine, his BPRS score had decreased to 49 (on haloperidol 5 mg BID and olanzapine 27.5 mg/day). On day 73 and day 47 of olanzapine 40 mg/day and propranolol 20 mg/day, L.'s BPRS score was 34. In general, L.'s positive symptoms improved, but his negative symptoms, including social isolation, alogia and social disengagement, persisted. L. also continued to pace around the unit.

Laboratory tests, including CBC, electrolytes, liver function tests, lipid panel, CPK, thyroid function tests, hepatitis panel, RPR, HIV, urinalysis, and urine toxicology screen, were unremarkable. Prolactin was mildly elevated at 24.2 ng/mL [normal: 4.6–21.4 ng/mL]. Head MRI demonstrated prominent ventricles and mild parenchymal volume loss, out of proportion to the patient's age. In addition, a cavum septum pellucidum and cavum vergae were observed. However, no acute hemorrhage, mass, mass effect, or other abnormal signals were seen.

During the hospital course, L. was followed by the cardiology service who confirmed that his cardiac function had returned to normal levels via quarterly ECG and biannual echocardiograms.

On hospital day 80 and olanzapine day 55, L. was discharged on olanzapine 40 mg/day and propranolol 20 mg/day to an adult unit at a different psychiatric hospital for further treatment.

Brief Formulation

In summary, L. is a 17-year-old bilingual Chinese adolescent boy with at least a three-year history of disorganized schizophrenia. Notable in his course of illness were persistent auditory hallucinations, violent and aggressive thoughts and behavior, illogical and disorganized speech, and history of marked cognitive decline. L. also experienced CIC.

At age 14 years, one year after L. moved to America, father reported decline in school performance and depressed mood, which was attributed to missing his mother who remained in China, and acculturation. At age 15 years, L. first appeared internally preoccupied and reported auditory hallucinations. He was minimally responsive to antipsychotic medication, although his family reports that he was compliant. He was intermittently hospitalized and discharged to day-treatment programs for two years, and subsequently transferred from an intermediate-care psychiatric hospital to a long-term state psychiatric hospital for treatment of visual and auditory hallucinations, bizarre delusions, thought disorder, and uncontrollable violent and aggressive outbursts. L.'s psychosis progressed insidiously and was precipitated by separation from his mother and relocation to the United States.

Multi-Axial Diagnoses

Discussion

This case illustrates the complexities of management of an apparently treatment-resistant adolescent with early-onset schizophrenia complicated by a major adverse effect of clozapine. L.'s presentation was consistent with CIC. His symptoms included hypotension, and tachycardia in the setting of ECG abnormalities and reduced cardiac function, which occurred as the dose of clozapine was being increased. The diagnosis was confirmed retrospectively, after clozapine had been discontinued and ventricular function returned to normal. At the time of diagnosis, L. was in his fourth month of clozapine treatment, and eighth day of further clozapine titration. This is consistent with reports results suggesting that cardiomyopathy is often known to present during clozapine dose titrations (Merrill et al. 2006).

L.'s treatment team initially did not suspect CIC, as his mild hypotension and tachycardia were attributed to benign orthostatic hypotension often associated with clozapine titration (Merrill et al. 2006; Cooper 2009). CIC is a rare condition that often presents with non-specific symptoms. The US Clozaril National Registry reported 41 cases in 189,405 adult patients who had taken clozapine from 1989 to 1999. In 22 of the 41 cases, cardiomyopathy was confirmed by signs and symptoms of heart failure, characteristic ECG changes or positive echocardiographic findings. The rate of CIC in children has not been established.

The mechanism by which clozapine leads to cardiomyopathy remains poorly understood. However, it has been speculated that clozapine induces myocarditis via direct injury or an immune-mediated hypersensitivity reaction, and that myocarditis may then progress to cardiomyopathy (Cooper 2009). As a result, myocarditis is a major risk factor for the onset of cardiomyopathy. Screening paradigms for clozapine-induced myocarditis have been suggested; however, such strategies remain inadequate since myocarditis is difficult to diagnose in both non-fulminant and fulminant forms. Patients with myocarditis may exhibit non-specific findings including fever, tachycardia, chest pain, dyspnea, flu-like symptoms, eosinophilia, elevated cardiac enzymes and ECG changes (Cooper 2009). In this case, L. experienced tachycardia, elevated CPK, mild hypotension, and mildly prolonged QTc. Hence, it is likely that L. also had myocarditis; however, this remains unclear because endocardial biopsy, the gold standard of diagnosis, was not performed.

L. may have been at risk because he was a young male with a pre-existing structural heart abnormality. Fortunately, because of L.'s history of mild aortic regurgitation, the team was diligent in pursuing a careful cardiac workup when symptoms of mild hypotension and tachycardia appeared. As a result, L.'s cardiomyopathy was diagnosed quickly with an echocardiogram. Early detection may have contributed to full recovery of cardiac function upon cessation of clozapine. Recovery from cardiomyopathy has also been seen in prior reports.

From 1969 to August 2001, a total of 178 cases of CIC were reported from multiple countries including the Canada, US, UK, Australia and others (Fontana and Sümegi 2002). Eighteen percent of the reported cardiomyopathy cases were fatal; however, in four cases where follow-up was reported, there was improvement of cardiac function upon withdrawal of clozapine. Other case reports also indicate improvement and/or full recovery with clozapine cessation (Merrill et al. 2006; Fontana and Sümegi 2002). Currently, the mainstay of treatment for CIC is cessation of clozapine and supportive care (Layland et al. 2009).

Despite the superior clinical efficacy of clozapine, its use in children and adolescents is limited by concern over the increased incidence of serious adverse effects. Here we demonstrate that, in addition to hematological monitoring, clinicians must also be aware of the potential for cardiotoxic effects, as younger patients are at risk for developing myocarditis and cardiomyopathy (Fontana and Sümegi 2002). Particular vigilance should be taken in monitoring patients who develop non-specific physical symptoms. A low threshold for concern and additional monitoring may be indicated during 1) the first several months of clozapine treatment, 2) in males, 3) in patients with pre-existing structural cardiac abnormalities, and 4) during any period of clozapine dose titration, as these groups could have an elevated risk for developing cardiomyopathy. Further investigation is needed to establish the most effective screening methods for clozapine-induced myocarditis and cardiomyopathy.

Treatment options following CIC include 1) re-challenge with clozapine, 2) switching to a different antipsychotic, and 3) electroconvulsive therapy (ECT). There are at least two case reports of patients who were successfully re-challenged with clozapine despite prior myocarditis (Floreani and Bastiampillai 2008; Rostagno et al. 2008). However, this approach is not generally recommended, as most re-challenge cases are associated with re-emergence of myocarditis (Lavland et al. 2009). A safer approach to treatment following CIC involves switching from clozapine to an antipsychotic agent with a lower risk of cardiotoxicity.

During L.'s course of treatment, clozapine was discontinued after cardiomyopathy was detected and he was initially maintained on haloperidol monotherapy. Although L.'s cardiomyopathy was attributed to clozapine, it is possible that clozapine and haloperidol acted synergistically to impair L.'s cardiac function, as both medications have been associated with myocarditis and cardiomyopathy (Coulter et al. 2001). Nevertheless, it soon became apparent that haloperidol monotherapy was ineffective in treating L.'s psychosis, and haloperidol was discontinued in favor of olanzapine.

The decision to re-challenge with olanzapine was made for several reasons, including evidence that 1) high-dose olanzapine can have equivalent efficacy to clozapine in adults with treatment-refractory schizophrenia (Meltzer et al. 2008; Volavka et al. 2002), 2) olanzapine can have beneficial effects in adolescents with treatment-refractory schizophrenia (Shaw et al. 2006; Kumra et al. 2008) and 3) L. had an incomplete initial trial with olanzapine at 10 mg/day.

L.'s psychiatric symptoms improved substantially on olanzapine 40 mg/day. His BPRS score decreased from 82 to 49. Because of aortic regurgitation and prior history of cardiomyopathy, L. was monitored carefully for the emergence of metabolic syndrome. At the time of transfer from the adolescent to the adult unit, L.'s cholesterol, triglycerides, HDL, LDL and fasting glucose were all within normal limits, and he had not experienced significant weight gain since starting olanzapine.

Compared to adults, children and adolescents may be at even greater risk for developing metabolic abnormalities with second generation antipsychotic use (Correll et al. 2009). L. showed minimal metabolic changes over the first seven weeks of olanzapine use. However, seven weeks may be too brief for adverse metabolic effects to emerge; Kumra et al. (2008) demonstrated increased hypertriglyceridemia and “pre-diabetes” (fasting blood glucose >100) by 24 weeks in 29% of children and adolescents who were treated with high-dose olanzapine.

Few other pharmacological options have been studied for treatment-refractory schizophrenia, especially in children and adolescents. Hence, the evidence base for other antipsychotic options is limited. However, there have been at least four double blind controlled trials investigating the efficacy of risperidone in adults with treatment-resistant schizophrenia. In the most recent and largest double blind trial to date in patients with treatment-resistant schizophrenia, risperidone was deemed effective, but improvement in psychotic symptoms was less robust when compared to clozapine (Azorin et al. 2001). In the case of L., since he had previously had an extended and ultimately unsuccessful trial of risperidone at an adequate dose (6 mg) and duration (11 months), high-dose olanzapine was felt to be the more appropriate choice following clozapine.

The final evidence-based treatment strategy for adults with treatment-resistant schizophrenia is ECT. A recent study found that patients with treatment refractory schizophrenia who received ECT in an open trial exhibited significant improvements in global pathology (Tang and Ungyari 2003). However, these improvements were not detected by measures of specific symptom domains. There is also some evidence suggesting that ECT may be helpful in patients who were partially responsive to antipsychotic medications. At the time of transfer to the adult hospital, L's BPRS score was 49 which represented a significant improvement from his pre-olanzapine baseline. Should L.'s symptoms worsen again while taking olanzapine, ECT might be considered as a potential option if pharmacotherapeutic alternatives have been exhausted.

In conclusion, to our knowledge, this represents the first case report to describe CIC in an adolescent, suggesting that clinicians need to be vigilant about this clinical complication in management of treatment refractory schizophrenia.