Clozapine

Hypertension

Hypertension has been associated with clozapine [ ], and alpha2-adrenoceptor blockade has been proposed as a possible mechanism [ ]. Four patients developed pseudopheochromocytoma syndrome associated with clozapine [ ]; all had hypertension, profuse sweating, and obesity. The authors suggested that clozapine could increase plasma noradrenaline concentrations by inhibiting presynaptic reuptake mediated by alpha2-adrenoceptors.

Hypotension

Hypotension is the most commonly observed cardiovascular adverse effect of neuroleptic drugs, particularly after administration of those that are also potent alpha-adrenoceptor antagonists, such as chlorpromazine, thioridazine, and clozapine [ ]. A central mechanism involving the vasomotor regulatory center may also contribute to the lowering of blood pressure.

• A 51-year-old man taking maintenance clozapine developed profound hypotension after cardiopulmonary bypass [ ].

Cardiac dysrhythmias

A substantial portion of patients taking clozapine develop electrocardiographic abnormalities; the prevalence was originally estimated at 10% [ ]. However, although the prevalence may be higher, most of the effects are benign and do not need treatment. In 61 patients with schizophrenia taking clozapine, in whom a retrospective chart review was conducted to identify electrocardiographic abnormalities, the prevalence of electrocardiographic abnormalities in those who used neuroleptic drugs other than clozapine was 14% (6/44), while in the neuroleptic drug-free patients it was 12% (2/17); when treatment was switched to clozapine, the prevalence of electrocardiographic abnormalities rose to 31% (19/61) [ ].

Tachycardia is the most common cardiovascular adverse effect of clozapine, and atrial fibrillation has also been reported [ , ]. The correlation between plasma clozapine concentration and heart rate variability has been studied in 40 patients with schizophrenia treated with clozapine 50–600 mg/day [ ]. The patients had reduced heart rate variability parameters, which correlated negatively with plasma clozapine concentration.

Clozapine can cause prolongation of the QT interval [ ].

• In a 30-year-old man taking clozapine there were minor electrocardiographic abnormalities, including a prolonged QT interval. A power spectrum analysis of heart rate variability showed marked abnormalities in autonomic nervous system activity. When olanzapine was substituted, power spectrum analysis studies showed that his heart rate had improved significantly and that his cardiovascular parameters had returned to normal. Serial electrocardiograms showed minimal prolongation of the QT interval.

The reports of sudden death associated with clozapine and the possibility that it may have direct prodysrhythmic properties have been reviewed [ ].

A patient developed ventricular fibrillation and atrial fibrillation after taking clozapine for 2 weeks [ ].

• A 44-year-old man with no significant cardiac history was given clozapine and 12 days later had bibasal crackles in the chest and ST segment elevation in leads V2 and V3 of the electrocardiogram. He then developed ventricular tachycardia and needed resuscitation. He also developed atrial fibrillation for 24 hours, which subsequently resolved.

Cardiomyopathy and myocarditis

Cardiomyopathy has been associated with neuroleptic drugs, including clozapine [ , , ], and partial data initially suggested an incidence of 1 in 500 in the first month.

• A 33-year-old woman without cardiac history or other susceptibility factors developed heart failure after taking clozapine and olanzapine for 6 weeks [ ]. After withdrawal of the two drugs, her heart function improved significantly.

• Atypical clozapine-induced cardiomyopathy occurred in a 28-year-old man who presented with a panic attack [ ].

• A 34-year-old man developed hypokinetic cardiomyopathy while taking clozapine and was successfully treated with carvedilol and captopril [ ].

The authors of the last case suggested that beta-blockers and ACE inhibitors may allow resumption of clozapine in refractory schizophrenia in which it has been withdrawn because of cardiotoxicity.

Myocarditis has been estimated to occur in 0.01–0.19% of patients taking clozapine [ ].

• A 32-year-old man developed myocarditis, documented by late gadolinium enhancement cardiovascular MRI, after taking clozapine for 1 week [ ]. The condition gradually resolved 5 weeks after withdrawal of clozapine.

Reports of fatal myocarditis have emerged.

• An 18-year-old man who had recently started to take clozapine (dose not specified) developed cardiac sounding chest pain, shortness of breath, and tachycardia of a few hours’ duration [ ]. Cardiac markers were raised and an electrocardiogram showed ST elevation in the inferior and lateral leads; echocardiography showed a dilated left ventricle and apical akinesia. Viral serology, rheumatoid antibody, thyroid function, autoimmune screen, and immunoglobulins were normal. When clozapine was withdrawn, he made a good recovery, with normalization of his cardiac markers and electrocardiogram.

• A 27-year-old man complained of flu-like symptoms, including generalized body aches, nasal congestion, and a scratchy throat 3 days after starting to take clozapine [ ]. He then developed slight dysarthria, fever, tachycardia, and an acute change in mental status. His serum troponin concentration was raised. An electrocardiogram showed sinus tachycardia with T wave inversion in lead III. Cardiac catheterization showed a markedly reduced left ventricular ejection fraction. Clozapine was withdrawn and he improved.

In a review of articles on adverse cardiac effects associated with clozapine, the estimated risk of potentially fatal myocarditis or cardiomyopathy was 0.01–0.19%; the authors suggested that this low risk of serious adverse cardiac events should be outweighed by a reduction in suicide risk in most patients [ ].

A thorough study of the risk of myocarditis or cardiomyopathy in Australia detected 23 cases (mean age 36 years; 20 men) out of 8000 patients treated with clozapine from January 1993 to March 1999 (absolute risk 0.29%; relative risk about 1000–2000) [ ]. All the accumulated data on previous reports of sudden death, myocarditis, or cardiac disease noted in connection with clozapine treatment were requested from the Adverse Drug Reactions Advisory Committee (ADRAC); there were 15 cases of myocarditis (five fatal) and 8 of cardiomyopathy (one fatal) associated with clozapine. All cases of myocarditis occurred within 3 weeks of starting clozapine. Cardiomyopathy was diagnosed up to 36 months after clozapine had been started. There were no confounding factors to account for cardiac illness. Necropsy results showed mainly eosinophilic infiltrates with myocytolysis, consistent with an acute drug reaction.

The manufacturers analysed 125 reports of myocarditis with clozapine and found 35 cases with fatal outcomes [ ]. A total of 53% occurred in the first month of therapy, and a small number (4.8%) occurred more than 2 years after the start of treatment. In this series, 70% of the patients were men.

Taking into account the results from an epidemiological study of deaths in users and former users of clozapine [ ], the cardiovascular mortality risk related to clozapine may be outweighed by the overall lower mortality risk associated with its beneficial effects, since the death rate was lower among current users (322 per 100 000 person years) than among past users (696 per 100 000 person years). The reduction in death rate during current use was largely accounted for by a reduction in the suicide rate compared with past use (RR = 0.25; CI = 0.10, 0.30).

Since cardiomyopathy is potentially fatal, some precautions must be taken. If patients taking clozapine present with flu-like symptoms, fever, myalgia, dizziness or faintness, chest pain, dyspnea, tachycardia or palpitation, and other signs or symptoms of heart failure, consideration should always be given to a diagnosis of myocarditis. Suspicion should be heightened if the symptoms develop during the first 6–8 weeks of therapy. It should be noted, however, that flu-like symptoms can also occur during the titration period, supposedly as a result of alpha-adrenoceptor antagonism by clozapine. Patients in whom myocarditis is suspected should be referred immediately to a cardiac unit for evaluation.

Clozapine rechallenge after myocarditis has been described [ ].

• A 23-year-old man with no history of cardiac disease was given clozapine 12.5 mg/day, increasing to 200 mg/day over 3 weeks; 5 weeks later he complained of shortness of breath and non-specific aches and pains in his legs and body. There was marked ST-segment depression and T wave inversion in the lateral and inferior leads of the electrocardiogram. There was no eosinophilia, and creatine kinase activity was not raised. An echocardiogram showed a hyperdynamic heart and left ventricular size was at the upper limit of normal. The heart valves were normal. Clozapine was withdrawn, but his mental state and quality of life deteriorated, and 2 years later clozapine was restarted because other drugs had not produced improvement. The dose of clozapine was built up to 225 mg at night and he remained well and free from cardiac adverse effects.

In this case a consultant cardiologist diagnosed myocarditis secondary to clozapine, as no other confounding co-morbidity was identified. However, the negative rechallenge suggests either that the clozapine was not responsible or that there was tolerance to the effect.

Since selenium is an essential antioxidant, and its deficiency has been implicated in myocarditis and cardiomyopathy, the aim of an observational study was to measure plasma and erythrocyte selenium concentrations in random venous blood samples from four groups: patients with mood disorders (n = 36), patients with schizophrenia taking clozapine (n = 54), patients with schizophrenia not taking clozapine (n = 41), and healthy controls (n = 56) [ ]. Selenium concentrations in plasma and erythrocytes were significantly lower in the patients taking clozapine compared with all the others. Thus, low selenium concentrations in patients taking clozapine may be important in the pathogenesis of life-threatening cardiac adverse effects associated with clozapine.

Pericarditis

Serositis (pericarditis and pericardial effusion, with or without pleural effusion) has been reported in patients taking clozapine [ ].

• A 43-year-old man developed a pericardial effusion after taking clozapine for 7 years. The condition resolved when the drug was withdrawn.

• A 21-year-old man with paranoid schizophrenia was treated with zuclopentixol, which was withdrawn because of extrapyramidal adverse effects, He was given clozapine 300 mg/day, and from day 43 developed breathlessness and complained of pain in his shoulders on deep inspiration. A chest X-ray showed an enlarged cardiac silhouette and bilateral pleural effusions. An echocardiogram showed pericardial and pleural effusions with no compromise of cardiac function. Clozapine was withdrawn and all the symptoms resolved within 2 weeks.

• A 33-year-old man developed pericarditis and a pericardial effusion after taking clozapine 200 mg bd for 10 weeks [ ]. The symptoms disappeared within 1 week after withdrawal of clozapine.

• A 47-year-old man developed an effuso-constrictive pericarditis associated with clozapine; his symptoms improved after drug withdrawal [ ].

• A 60-year-old woman developed polyserositis with a pericardial effusion and bilateral pleural effusions after taking clozapine for 13 months [ ].

• A 43-year-old woman developed clozapine-associated polyserositis (pleuritis and serositis) and hepatitis, although an electrocardiogram and a transthoracic echocardiogram were normal [ ].

• A 16-year-old girl developed pericarditis associated with clozapine. There were electrocardiographic changes and serial rises in serum troponin I, a highly sensitive and specific marker of myocardial injury.

In the last case there were rises in troponin I, which resolved despite continuation of therapy. The authors suggested that troponin I is the preferred marker for monitoring the cardiac adverse effects of clozapine.

Venous thromboembolism

Typical neuroleptic drugs have been associated with an increased risk of venous thromboembolism [ ]. Data from the Swedish Reactions Advisory Committee suggested that clozapine is also associated with venous thromboembolic complications [ ]. Between 1 April 1989 and 1 March 2000, 12 cases of venous thromboembolism were collected; in 5 the outcome was fatal. Symptoms occurred in the first 3 months of treatment in eight patients; the mean clozapine dose was 277 mg/day (75–500). Although during the study total neuroleptic drug sales, excluding clozapine, accounted for 96% of all neuroleptic drug sales, only three cases of thromboembolism associated with those neuroleptic drugs were reported. The reported risk of thromboembolism associated with clozapine is estimated to be 1 per 2000–6000 treated patients, the true risk being higher owing to under-reporting. These conclusions were consistent with those from an observational study [ ].

Anecdotal reports have also appeared [ ].

Between February 1990, when clozapine was first marketed in the USA, and December 1999 the FDA received 99 reports of venous thromboembolism (83 mentioned pulmonary embolism with or without deep vein thrombosis and 16 mentioned deep vein thrombosis alone) [ ]. In 63 cases death had resulted from pulmonary embolism; 32 were confirmed by necropsy. Of 36 non-fatal cases, only 7 had been documented objectively by such diagnostic techniques as perfusion-ventilation lung scanning and venography. Thus, in 39 of the 99 reports there was objective evidence of pulmonary embolism or deep vein thrombosis. The median age of the 39 individuals was 38 (range 17–70) years and 20 were women. The median daily dose was 400 (range 125–900) mg. The median duration of clozapine exposure before diagnosis was 3 months (range 2 days to 6 years). Information on risk factors for pulmonary embolism and deep vein thrombosis varied; however, 18 of the 39 patients were obese. The frequency of fatal pulmonary embolism in this study is consistent with that described in the labelling for clozapine in the USA.

As of 31 December 1993, there were 18 cases of fatal pulmonary embolism in association with clozapine therapy in users aged 10–54 years. Based on the extent of use recorded in the Clozapine National Registry, the mortality rate associated with pulmonary embolism was 1 death per 3450 person years of use. This rate was about 28 times higher than that in the general population of a similar age and sex (95% CI = 17, 42). Whether pulmonary embolism can be attributed to clozapine or some characteristic(s) of its users is not clear [ ].

Fatal pulmonary embolism occurred in a 29-year-old man who was not obese, did not smoke, and had not had recent surgery, after he had taken clozapine 300 mg/day for 6 weeks [ ].

• A 58-year-old white deaf man, with a history of pulmonary embolism, two first-degree relatives with a history of stroke and myocardial infarction, and one first-degree relative who died suddenly, developed a new episode of pulmonary embolism shortly after clozapine was begun [ ].

The authors pointed out that this case suggests that susceptibility factors, such as a previous history of pulmonary embolism or venous thrombosis or a strong family history, could be viewed as relative contraindications to treatment with clozapine; however, in an invited comment it was pointed out that in this case it was not mentioned whether the patient was immobile or not, because a crucial risk factor for thrombosis in psychiatric patients is reduced physical activity [ ].

In another case venous thromboembolism occurred on two occasions in a 22-year-old man taking clozapine [ ]. Resolution of the first episode was most probably due to treatment with anticoagulants, and the authors thought that withdrawal of anticoagulants could be regarded as equivalent to reintroduction of clozapine. They suggested that, although the interval between the withdrawal of anticoagulant therapy and the second episode was long (26 months), the sequence of events suggested a causal relation between clozapine and venous thromboembolism.

Sleep

In a study of the effects of clozapine on the electroencephalogram, 13% of patients developed spikes with no relation to dose or serum concentration of clozapine; 53% developed electroencephalographic slowing. Compared with plasma concentrations below 300 ng/ml, a clozapine serum concentration of 350–450 ng/ml led to more frequent and more severe electroencephalographic slowing [ ]. There were considerable differences in the electroencephalographic patterns between classical neuroleptic drugs and clozapine [ ]. Clozapine-treated patients showed significantly more stage 2 sleep, more stable non-REM sleep (stages 2, 3, and 4), and less stage 1 than patients treated with haloperidol or flupentixol. In a longitudinal study, clozapine significantly improved sleep continuity and significantly increased REM density, but did not affect the amount of REM sleep [ ].

Stuttering

Clozapine can cause stuttering [ ]. The pathogenesis of developmental stuttering, as well as acquired or neurogenic stuttering, is unclear. However, since clozapine-induced stuttering can precede a seizure it may be related to an effect on the brain rather than to a dystonic syndrome, as previously suggested.

• A 28-year-old man taking clozapine 300 mg/day developed severe stuttering and subsequently had a generalized tonic-clonic seizure while taking 425 mg/day [ ]. There were electroencephalographic abnormalities, especially left-sided slowing.

• A 57-year-old man developed severe stuttering after taking clozapine 300 mg/day and sodium valproate 600 mg/day for 8 months [ ]. It abated 1 week after clozapine was reduced and finally withdrawn.

• A 49-year-old woman had prominent stuttering before a generalized epileptic seizure and recovered after antiepileptic treatment [ ].

• A 40-year-old man started to stutter when his clozapine dosage was increased from 400 mg/day; it was also associated with a marked increase in seizure activity [ ].

• A 44-year-old man developed a clozapine-induced speech dysfluency, which disappeared when clozapine was withdrawn [ ].

Electroencephalographic abnormalities and stuttering may be harbingers of seizures and clozapine is thought to increase the risk of this [ ]. A 74-year-old man and a 56-year-old man, both with non-epileptic clozapine-induced drop attacks, developed bouts of sudden collapse without loss of consciousness, and symptoms of posterior circulation disruption [ ].

Seizures

Clozapine has a proconvulsant effect. Factors that increase the likelihood of seizures include high doses of clozapine, rapid dose titration, the concurrent use of other epileptogenic agents (such as antidepressants, neuroleptic drugs, and mood stabilizers) and a previous history of neurological abnormalities [ ].

The prevalence of seizures with clozapine is higher than average (about 5%) and is dose-dependent [ , ].

• A tonic-clonic seizure occurred in a 30-year-old man 4 weeks after he started to take clozapine 400 mg/day [ ]. This was followed by a large increase in liver enzymes, which had been normal the week before.

Seizure characteristics and electroencephalographic abnormalities in 12 patients taking clozapine have been identified; there was a surprisingly high incidence of focal epileptiform abnormalities [ ]. Seizures associated with clozapine are dose-dependent [ ]. However, there have been reports of seizure activity in patients taking therapeutic or subtherapeutic doses of clozapine [ , ]. Seizures have occasionally been reported in patients taking low doses.

• A 28-year-old woman, with no history of prior seizures and not taking concomitant medication, had seizures while taking clozapine 200 mg/day [ ].

• A 75-year-old patient developed seizures while taking clozapine 12.5 mg/day.

However, in the second case the seizure was unlikely to have been due to clozapine, given the very low dose and non-recurrence with rechallenge at higher dosages [ ].

Despite the risk of seizures in patients without pre-existing epilepsy, six patients with epilepsy and severe psychosis taking clozapine had no increases in seizure frequency, and three had a substantial reduction [ ].

Several anticonvulsants have been shown to be helpful in the prevention and treatment of clozapine-induced seizures.

• A 15-year-old boy with refractory schizophrenia had seizures with clozapine; he was given gabapentin, and several years later was free of seizures [ ].

The addition of lamotrigine to clozapine therapy has been associated with rapid improvement of psychiatric symptoms [ ]; this has been observed in three cases of poor response or resistance to clozapine monotherapy.

Cataplexy

Cataplexy, an abnormal state characterized by atonia and believed to represent dissociated REM sleep phenomena that intrude into wakefulness, has been reported in a 29-year-old woman taking clozapine [ ].

Extrapyramidal effects

Clozapine has a more favorable extrapyramidal effects profile than other neuroleptic drugs [ ] and little or no parkinsonian effect [ ].

Akathisia

The low prevalence of akathisia in patients taking clozapine has led to the proposal that clozapine should be used to treat patients with neuroleptic drug-induced chronic akathisia [ , ]. However, a case of acute nocturnal akathisia has been reported in a 43-year-old man taking clozapine who was successfully treated with beta-blockers [ ].

Parkinsonism

The efficacy and safety of treatment with clozapine in patients with Parkinson’s disease have been discussed [ ], and a multicenter retrospective review of the effects of clozapine in 172 patients with Parkinson’s disease has been published [ ]. The mean duration of clozapine treatment was 17 (range 1–76) months. Low-dose clozapine improved the symptoms of psychosis, anxiety, depression, hypersexuality, sleep disturbances, and akathisia. Of the 40 patients, 24% withdrew as a result of adverse events, mostly sedation (n = 19). Sedation was reported in 46%, sialorrhea in 11%, and postural hypotension in 9.9%. Neutropenia was detected in four patients (2.3%).

Six patients who met the criteria for a diagnosis of HIV-associated psychosis, and who had previously developed moderate parkinsonism as a result of typical neuroleptic drugs, were treated with clozapine [ ]. Parkinsonism improved by an average of 77%, but one patient did not complete the trial because of a progressive fall in leukocyte count.

Clozapine has been used to treat psychosis related to Parkinson’s disease [ ]. In a randomized, double-blind, placebo-controlled trial of low doses of clozapine (6.25–50 mg/day) in 60 patients (mean age 72 years) with idiopathic Parkinson’s disease and drug-induced psychosis, the patients in the clozapine group had significantly more improvement after 14 months than those in the placebo group in all measures used to determine the severity of psychosis [ ]. Clozapine improved tremor and had no deleterious effect on the severity of parkinsonism, but in one patient it was withdrawn because of leukopenia.

In a randomized, double-blind, placebo-controlled, 4-week trial in 60 patients with similar drug-induced psychosis in Parkinson’s disease, assigned to clozapine (n = 32) or placebo (n = 28), the initial clozapine dose of 6.25 mg/day was titrated over at least 10 days to a maximum of 50 mg/day and was rapidly effective [ ]. Somnolence and worsening of parkinsonism were significantly more frequent in the clozapine group, seven of whom reported worsening of Parkinson’s disease, usually mild or transient, which was confirmed by aggravation of the Schwab and England score by 10–20% in three patients; however, no-one withdrew for this reason.

Of 32 patients with Parkinson’s disease and psychosis (mean age 73 years; mean disease duration 12.3 years) who were followed for 5 years in a non-randomized, open study 19 (eight with dementia) continued to take clozapine (mean dose 50 mg/day) and 13 stopped taking it [ ]. The average duration of treatment in those in whom medication was stopped was 8.5 (range 1–24) months; the reasons for withdrawal were: symptoms improved and did not return after weaning off clozapine (n = 9), somnolence (n = 3), and personal reasons (n = 1). There was no correlation between age, sex, duration and severity of disease, the presence of dementia, and the response to clozapine. Also, the Parkinsonian Psychosis Rating Scale scoring did not influence clozapine response.

Restless legs syndrome is a neurological disorder characterized by irresistible movements and dysesthetic sensations in the legs. A case associated with clozapine has been described, supposedly for the first time [ ].

• A 26-year-old man developed severe extrapyramidal symptoms characterized by tremor, bradykinesia, and sialorrhea while taking haloperidol 20 mg/day and valproate 1400 mg/day. Haloperidol was replaced by clozapine 50 mg/day and valproate was continued. After 3 days he started to have unpleasant sensations in his calves and burning sensations over his feet.

Tardive dyskinesia

It is said that clozapine causes less tardive dyskinesia than haloperidol and even that it can improve pre-existing tardive dyskinesia [ ].

Clozapine-associated tardive dyskinesia and hypothyroidism has been reported [ ].

• A 47-year-old woman currently taking levothyroxine for hypothyroidism took clozapine for schizophrenia and after 7 months developed horizontal grinding movements of the lower jaw and dyskinetic movements of the tongue. She scored 9 on the Abnormal Involuntary Movements Scale (AIMS). Levothyroxine was withdrawn for 8 weeks without any effect on the abnormal movements or the AIMS score; neither a reduction nor an increase in the dose of clozapine produced any improvement in the dyskinetic movements.

Patients with schizophrenia with (n = 15) and without (n = 11) tardive dyskinesia differed markedly in their dopaminergic response to haloperidol, assessed by means of plasma homovanillic acid variations, which increased, whereas this difference was not observed after clozapine [ ].

Nevertheless, 46 patients taking clozapine had higher tardive dyskinesia scores compared with 127 taking typical neuroleptic drugs [ ]. In a multiple regression analysis, there was a significant relation between the total score on the Abnormal Involuntary Movement Scale (AIMS) as a dependent variable and current neuroleptic drug dose, duration of treatment, age, sex, diagnosis, current antiparkinsonian therapy, and illness duration. There was no beneficial effect of clozapine on the prevalence of tardive dyskinesia, and the authors’ conclusion was that certain patients develop tardive dyskinesia despite long-term intensive clozapine treatment; however, since most clozapine users were past users of typical neuroleptic drugs, this conclusion must be regarded with caution.

• A 45-year-old woman developed tardive dyskinesia while taking clozapine [ ]. She had never had any of the symptoms before she started to take 223 mg/day and first experienced involuntary tongue movements and akathisia 5 months after the start of treatment.

• Tardive dyskinesia has been attributed to clozapine in a 44-year-old man, who had discontinued haloperidol 24 days before the event [ ].

Clozapine has been evaluated in an open study in seven patients (mean age 29 years; mean dose 428 mg/day) with chronic exacerbated schizophrenia and severe tardive dyskinesia [ ]. Extrapyramidal Symptoms Rating Scale scores fell by 83% after 3 years and 88% after 5 years. None of the patients had adverse effects related to clozapine: their weight did not change significantly and their serum glucose, cholesterol, and triglyceride concentrations remained within the reference ranges.

Tardive dystonia

It is generally considered that clozapine has little or no potential to cause tardive dystonia; it has even been speculated that it may be an effective therapy for this adverse effect [ ]. The efficacy of clozapine in severe dystonia was therefore assessed in an open trial in five patients [ ]. All had significant improvement; nevertheless, all had adverse reactions, such as sedation and orthostatic hypotension; in one case persistent symptomatic orthostatic hypotension and tachycardia limited treatment. However, there was no evidence of a beneficial effect of clozapine in primary dystonia, the most common form of dystonia and a difficult disorder to treat, until the report of a 56-year-old woman with severe and persistent primary cranial dystonia (Meige’s syndrome), who responded to clozapine (50–100 mg) [ ]. Two cases of severe oromandibular dystonia refractory to other antidystonic therapies, including botulinum toxin, which improved with clozapine have also been reported [ ].

Nevertheless there have been reports of dystonias associated with clozapine.

• A 37-year-old man, who had taken numerous neuroleptic drugs from 1975 to 1990, was switched to clozapine because of breakthrough psychosis [ ]. Clozapine was effective and was his only neuroleptic drug treatment from that time. In 1996, when his clozapine dosage was 825 mg/day, he had left torticollis of 60–70°, mild left laterocollis, and superimposed spasmodic head movements jerking his head to the left. He had difficulty rotating his head to the right past the midline.

• Dystonia in a 34-year-old man associated with clozapine satisfied the criteria for antipsychotic drug-induced tardive dyskinesia [ ].

There has also been a report of blepharospasm, a type of focal tardive dystonia that is usually considered to be a variant of tardive dyskinesia, in a 46-year-old woman [ ]. On the other hand, an antidyskinetic effect of zotepine on clozapine-associated tardive dyskinesia has been reported [ ].

Tardive tremor

Tardive tremor is a hyperkinetic movement disorder associated with chronic neuroleptic drug treatment. It was first described in 1991 as a symmetrical tremor, of low frequency, present at rest and during voluntary movements but most prominent during maintenance of posture, and often accompanied by tardive dyskinesia. Tetrabenazine is the current treatment. Sequential responsiveness to both tetrabenazine and clozapine has been reported [ ].

• A 55-year-old man with a 15-year history of schizophrenia treated with various neuroleptic drugs developed a tremor and was given tetrabenazine 75 mg/day, with complete regression of the tremor. Three months later he developed depression, a known adverse effect of tetrabenazine, which was discontinued, with subsequent partial improvement of his depressive symptoms but reappearance of the tardive tremor. Clozapine 25 mg/day was started and increased to 75 mg/day; his tardive tremor again disappeared.

Neuroleptic malignant syndrome

Neuroleptic malignant syndrome has been associated with clozapine [ , ], although some doubts were expressed about the features of earlier cases. However, atypical features have been described [ ]. In the light of two cases, a 35-year-old man and a 62-year-old woman, the literature was comprehensively reviewed and the characteristics of neuroleptic malignant syndrome due to clozapine and typical neuroleptic drugs were compared [ ]. Causation with clozapine was deemed highly probable in 14 cases, of medium probability in 5 cases, and of low probability in 8 cases. The most commonly reported clinical features were tachycardia, changes in mental status, and sweating. Fever, rigidity, and raised creatine kinase activity were less prominent than in the neuroleptic malignant syndrome associated with typical neuroleptic drugs. This suggests that the presentation of clozapine-induced neuroleptic malignant syndrome may be different from that of typical neuroleptic drugs. Two other cases have also illustrated that possibility [ , ].

Neuroleptic malignant syndrome and subsequent acute interstitial nephritis has been reported in a 44-year-old woman [ ]. She met the main criteria for neuroleptic malignant syndrome, although she did not develop rigidity or a rise in creatine kinase activity. On the other hand, abnormal creatine kinase activity and signs of myotoxicity were respectively found in 14% and 2.1% of patients who took clozapine for an average of 18 months (n = 94) [ ].

• A 22-year-old man developed atypical neuroleptic malignant syndrome while taking clozapine [ ]. He vomited and was sweating and agitated but afebrile, with mild hypertension (maximum 156/96 mmHg) and a tachycardia, with marked increases in white blood cell count (32 × 109/l), neutrophils (25 × 109/l), and creatine kinase (1442 IU/l); a similar syndrome occurred while he was taking haloperidol.

• A 52-year-old man with risk factors, including a subdural hematoma and three prior episodes of neuroleptic malignant syndrome secondary to chlorpromazine, loxapine, and lithium, developed neuroleptic malignant syndrome while taking clozapine [ ].

Hyperlipidemia

Metabolic syndrome is a collection of risk factors that are associated with increased morbidity and mortality due to cardiovascular disease. Since clozapine has been associated with a substantial effect on metabolic parameters, such as weight gain, hypertriglyceridemia, increased total cholesterol concentrations, and hypertension, the aim of a controlled open study was to compare the prevalence of the metabolic syndrome among 93 out-patients and a matched group of 2701 subjects drawn from a database [ ]. The metabolic syndrome was defined as three or more of the following: a waist circumference over 102 cm for men and over 88 cm for women; a fasting blood triglyceride concentration higher than 1.7 mmol/l (150 mg/dl); a high-density lipoprotein cholesterol concentration below 1.0 mmol/l (40 mg/dl) for men and 1.3 mmol/l (50 mg/dl) for women; a blood pressure of over 130 mmHg systolic or 85 mmHg diastolic; and a fasting blood glucose concentration over 5.6 mmol/l (100 mg/dl). The prevalence of the metabolic syndrome was significantly higher among patients taking clozapine (54%) than in the comparison group (21%); a logistic regression analysis showed significant associations with age and body mass index in both groups, and treatment duration for clozapine.

• A 42-year-old man with a schizoaffective disorder had new-onset hyperlipidemia while taking clozapine (after failing therapy with traditional antipsychotic drugs) [ ]. Before taking clozapine his total cholesterol measurements were 2.9–5.5 mmol/l and there were no triglyceride measurements. Despite treatment with various antihyperlipidemic agents, his total cholesterol concentration reached 12 mmol/l and his triglyceride concentration reached 54 mmol/l. His antipsychotic drug therapy was switched to aripiprazole and his lipid concentrations improved dramatically, to the point that antihyperlipidemic treatment was withdrawn. When he was given clozapine again his lipid concentrations again worsened.

Diabetes mellitus

Several cases of de novo diabetes mellitus or exacerbation of existing diabetes in patients taking neuroleptic drugs have been reported, including patients taking clozapine [ ]. There was no significant relation to weight gain.

• A 49-year-old man taking olanzapine developed diabetes mellitus and recovered after withdrawal [ ].

• Diabetic ketoacidosis occurred in a 31-year-old man who had taken clozapine 200 mg/day for 3 months for refractory schizophrenia [ ]. Clozapine was withdrawn and he remained metabolically stable. Two months later, clozapine was restarted, and only 72 hours after drug re-exposure he had increased fasting glycemia and insulinemia, suggesting insulin resistance as the underlying mechanism. Apart from slight obesity, he had no predisposing factors.

Hyperglycemia occurs at 2 weeks to 3 months after the start of clozapine treatment and occurs without predisposing factors. Clozapine-induced hyperglycemia can be serious, leading to coma, but it is reversible if clozapine is withdrawn. In some cases, continuation of clozapine is possible by controlling blood glucose concentrations with hypoglycemic drugs. This approach can be useful in refractory schizophrenia responsive to clozapine. All patients should be advised to report altered consciousness, polyuria, or increased thirst.

Glucose metabolism has been studied in 17 patients taking clozapine [ ]. Six had impaired glucose tolerance and eight had a glycemic peak delay.

Diabetes was common in 63 patients taking clozapine than in 67 receiving typical depot neuroleptic drugs [ ]. The percentages of type 2 diabetes mellitus were 12% and 6% respectively. Nevertheless, the mechanism is not known. In six patients with schizophrenia, clozapine increased mean concentrations of blood glucose, insulin, and C peptide [ ]. The authors concluded that the glucose intolerance was due to increased insulin resistance.

However, opposite data have been found in a case–control study in 7227 patients with new diabetes and 6780 controls, all with psychiatric disorders [ ]. Clozapine was not significantly associated with diabetes (adjusted OR = 0.98; 95% CI = 0.74, 1.31) and there was no suggestion of relations between larger dosages or longer durations of clozapine use and an increased risk of diabetes. Among individual non-clozapine neuroleptic drugs, there were significantly increased risks for two phenothiazines: chlorpromazine (OR = 1.31; 95% CI = 1.09, 1.56) and perphenazine (OR = 1.34; 95% CI = 1.11, 1.62). The authors suggested that, in contrast to earlier reports, these results provided some reassurance that clozapine does not increase the risk of diabetes. However, cases of diabetes were identified by the new use of antidiabetic drugs, and it is therefore possible that clozapine was associated with less pronounced glucose intolerance that did not require drug therapy.

The effect of clozapine on glucose control and insulin sensitivity has been studied prospectively in 9 women and 11 men with schizophrenia (mean age 31 years) [ ]. Insulin resistance at baseline was unaffected by clozapine, but 11 of the patients developed abnormal glucose control (mean age 30 years; five women). Mean fasting and 2-hour glucose concentrations increased significantly by 0.55 mmol/l. There was no correlation between change in body mass index and change in fasting glucose concentrations. Weight gain with clozapine compared with other neuroleptic drugs has been studied in in-patients who were randomly assigned to switch to open treatment with clozapine (n = 138) or to continue receiving conventional neuroleptic drugs (n = 89) [ ]. Patients gained weight at the end of 2 years whether they switched to clozapine (5.9 kg, 7%) or continued to take first-generation neuroleptic drugs (2.3 kg, 4%), but weight gain was significantly greater (1 body mass index unit) in those taking clozapine, particularly women.

Weight gain

Weight gain is often associated with clozapine [ ]. In 42 patients who took clozapine for at least 1 year, men and women gained both weight and body mass, which is more directly related to cardiovascular morbidity [ ]. Over 10 weeks, leptin concentrations, which correlate with body mass index, increased significantly from baseline in 12 patients taking clozapine [ ].

An 8-year retrospective chart review of 96 hospitalized patients with schizophrenia has analysed data on monthly weight change, initial response, age, sex, clozapine dose, and concomitant use of mood stabilizers and other antipsychotic drugs [ ]. There was an average weight gain of 11.7 kg in 55 patients who took clozapine over the entire 8-year period; of these, 17 who had a significant initial clinical response gained significantly more weight (13.8 kg) than the 38 patients without a significant initial response (4.5 kg). Lower baseline body mass index was also associated with significantly more weight gain.

The relation between genetic variants of the β3 adrenoceptor and the G-protein β3 subunit and clozapine-induced body weight change has been investigated in 87 treatment-resistant patients with schizophrenia [ ]. They gained an average of 2.6 kg. There was no statistically significant relation between weight gain and either the β3 adrenoceptor Trp6Arg or the G-protein β3 subunit C8257 polymorphisms.

In a long-term follow-up study (14 months) of 93 patients with schizophrenia the possible relation between clozapine-induced weight gain and a genetic polymorphism in the adrenoceptor α2a receptor, − 1291 C > G, has been examined [ ]. The GG genotype was associated with a significantly higher mean body weight gain (8.4 kg) than the CC genotype (2.8 kg).

The effect of clozapine on serum ghrelin concentrations has been investigated in 12 patients over 10 weeks after the start of treatment [ ]. In contrast to increased body mass indices and serum leptin concentrations, there were no significant changes in serum ghrelin concentrations. The authors claimed that these results do not support a causal involvement of ghrelin in clozapine-related weight gain.

The possible variables associated with weight gain have been analysed in 50 treatment-refractory patients with schizophrenia who were randomized to clozapine 100, 300, or 600 mg/day in a 16-week, double blind study [ ]. Weight gain varied across three baseline body mass index (BMI) categories: normal weight (+ 4.1 kg), overweight (+ 2.6 kg) and obese (+ 0.4 kg), and according to dosing: 600 mg (+ 4.4 kg), 300 mg (+ 2.6 kg), and 100 mg (+ 1.3 kg). Sex had no effect after controlling for baseline BMI and dose, but African–American ethnicity had a strong significant effect, despite the small number included in the sample (n = 6). Plasma norclozapine concentration was not significantly correlated with weight gain.

The association between clozapine-related weight gain and increased mean arterial blood pressure has been examined in 61 patients who were randomly assigned to either clozapine or haloperidol in a 10-week parallel-group, double-blind study, and in 55 patients who chose to continue to take clozapine in a subsequent 1-year open study [ ]. Clozapine was associated with significant weight gain in both the double-blind trial (mean 4.2 kg) and the open trial (mean 5.8 kg). There was no significant correlation between change in weight and change in mean arterial blood pressure.

There were no significant associations between cycle length and weight change during clozapine treatment in 13 premenopausal women with psychoses [ ].

Sleep apnea associated with clozapine-induced obesity has been reported [ ].

• A 45-year-old woman with schizophrenia who took clozapine 300 mg/day for 16 months gained 18 kg and had hypertriglyceridemia and glucose intolerance. She had daytime sedation, difficulty in sleeping at night, loud snoring, and periods of apnea during sleep. Nasal continuous positive airway pressure produced improvement.

Phenylpropanolamine 75 mg/day did not promote weight loss in a randomized, placebo-controlled study in 16 patients with schizophrenia who had gained at least 10% of their body weight while taking clozapine [ ]).

• A 29-year-old man taking clozapine 800 mg/day gained 46 kg in weight after 25 months, and had myoclonic jerks in the hands, arms, and shoulders on both sides [ ]. He was treated with topiramate (which causes weight loss). The myoclonic jerks disappeared completely. He lost 21 kg over 5 months, with no significant change in eating habits or food consumption, and felt more energetic, more active, and more motivated to exercise.

• The 22-year-old son of healthy parents, with a life-long history of galactosemia, developed weight gain while taking an effective dose of clozapine [ ].

Because patients with galactosemia need to avoid weight loss as a result of restrictive dietary measures, the authors suggested that this is an interesting example of weight gain as a positive side effect of clozapine, not necessarily associated with increased appetite and higher caloric intake.

Neutropenia and agranulocytosis

Incidence : The incidence of clozapine-induced agranulocytosis was originally determined to be 0.21% in a selected Finnish population [ ], and the drug was withdrawn, only to be cautiously reintroduced in some countries a decade later, with hematological monitoring. With mandatory hematological monitoring by the Clozaril Patient Management System in the USA, the cumulative incidence of agranulocytosis was 0.8% at 1 year and 0.9% at 1.5 years of treatment [ ]. In France, the incidences of agranulocytosis and neutropenia in clozapine-treated patients from December 1991 were 0.46 and 2.1% respectively [ ]. Some of the available postmarketing data on clozapine-induced agranulocytosis are presented in Table 1 [ ]. The EIDOS and DoTS descriptions are shown in Figure 1 .

Agranulocytosis can occur several months or even years after withdrawal of clozapine.

• A 49-year-old mentally retarded man took clozapine for 7 years and after suffering from severe recurrent infections developed fatal agranulocytosis more than 4 years after discontinuing clozapine [ ].

Mechanism : The underlying mechanisms of agranulocytosis are unknown, but hemopoietic cytokines, such as granulocyte colony-stimulating factor (G-CSF), are likely to be involved [ ].

• In a 26-year-old woman who developed granulocytopenia twice, first when taking clozapine and again when taking olanzapine, G-CSF concentrations, but not those of other cytokines, closely paralleled the granulocyte count.

• In a 73-year-old patient who developed granulocytopenia while taking clozapine, G-CSF and leukocyte counts were reliable indicators of the evolution of the condition, showing an abortive form of toxic bone-marrow damage with subsequent recovery [ ].

Immune-mediated mechanisms of clozapine-induced agranulocytosis have been reviewed in the context of agranulocytosis in a 46-year-old woman [ ]. In three patients who developed agranulocytosis, seven who developed neutropenia, and five who were asymptomatic, there was no evidence of antineutrophil antibodies in the blood shortly after an episode of clozapine-induced agranulocytosis, and an antibody mechanism seems unlikely, in view of the delay in onset of clozapine-induced agranulocytosis on re-exposure to the drug [ ].

• Increased apoptosis of neutrophils has been reported in a 45-year-old woman with clozapine-induced agranulocytosis [ ]. Withdrawal of clozapine and treatment with granulocyte colony-stimulating factor led to normalization of the blood neutrophil count within 3 weeks.

The authors suggested that enhanced apoptosis of blood neutrophils during the acute phase of clozapine-induced agranulocytosis could have resulted from enhanced expression of the pro-apoptotic proteins Bax and Bilk and from a reduction in the anti-apoptotic proteins BCI-XLmRNA. The time-course of the fall and recovery of the neutrophils, as well as the release pattern of endogenous G-CSF, resembled those of chemotherapy-induced neutropenia. The kinetics of CD 34-positive cells mimic those of cytotoxic progenitor cell mobilization, for example after cytostatic drug administration. They suggested that clozapine-mediated inhibition of release of G-CSF or granulocyte-macrophage colony-stimulation factor (GM-CSF) is involved in clozapine-induced agranulocytosis.

Susceptibility factors : Some of the genetic aspects of clozapine-induced agranulocytosis have been evaluated [ ]. Polymorphisms of specific clozapine metabolizing enzyme systems were determined in 31 patients with agranulocytosis and in 77 without. Genotyping of a recently discovered G-463 A polymorphism of the myeloperoxidase gene and CYP2D6 showed no evidence of an association.

Because of the unusually high incidence of agranulocytosis in Finnish and Jewish patients [ ], an ethnic susceptibility factor for agranulocytosis has been suggested. Human leukocyte antigen (HLA) B38 phenotype was found in 83% of patients who developed agranulocytosis and in 20% of clozapine-treated patients who did not develop agranulocytosis [ ]. Gene products contained in the haplotype may be involved. In an open study in 31 German patients with clozapine-induced agranulocytosis and 77 controls with schizophrenia, agranulocytosis was significantly associated with HLA-Cw*7, DQB*0502, DRB1*0101, and DRB3*020 [ ]. No other antigens were associated with agranulocytosis, but age was another major susceptibility factor. In another study in two groups of Finnish patients (19 “clozapine responders” and 26 patients with a history of non-fatal clozapine-induced granulocytopenia or agranulocytosis), the frequency of the HLA-A1 allele in the latter was low (12%), whereas HLA-A1 was associated with a good therapeutic response at an allele frequency of 58% (the frequency of HLA-A1 being 20% in the Finnish population) [ ].

Concordant clozapine-induced agranulocytosis in monozygotic twins also suggested a genetic susceptibility; in both twins there was a low leukocyte count after 9 weeks of treatment [ ]. Serological typing of the HLA system showed identical patterns in the twins: HLA-A: 28, 26; HLA-B: 49, 63; DR: 2 (versus 16), 12, 52; DQ: 1. The authors pointed out that these data suggest that genetic factors may participate not only in the time of onset of schizophrenia, but also in the emergence and timing of agranulocytosis in response to clozapine.

Rates of leukopenia and agranulocytosis as reasons for withdrawal of clozapine have been examined retrospectively in 1875 patients who took clozapine between 1989 and 1999 [ ]. No African–American patients developed agranulocytosis, while eight Caucasian patients (0.62%) did; however, 5.3% of the African–American cohort versus 2.4% of the Caucasian cohort had to stop taking clozapine treatment because of leukopenia. The authors suggested that it is likely that the African–American patients had clozapine withdrawn unnecessarily because of benign ethnic neutropenia, within a lower reference range for white blood cell count.

Clinical features : Careful attention should be paid to possible early warnings of agranulocytosis, such as fever, sore throat, and lymphadenopathy.

Circadian variation in white cell count, with a dip in the morning, has been misdiagnosed as clozapine-induced neutropenia [ ].

• A 31-year-old man with resistant schizophrenia took clozapine 500 mg/day. Although this was effective, the granulocyte count fell to 1.2 × 109/l (total count not given) and clozapine was withdrawn. During the subsequent year, several neuroleptic drugs were used, with unsatisfactory results. Careful monitoring showed a pronounced diurnal variation in both total white cell count (2.9–4.2 × 109/l in the morning and 3.6–7.1 × 109/l in the afternoon) and granulocytes (0.8–1.4 × 109/l in the morning and 2.9–5.5 × 109/l in the afternoon).

Thus, an apparently low white cell count may simply reflect the nadir of the diurnal variation and may not indicate a need to withdraw clozapine.

In over 11 000 patients the risk of agranulocytosis was higher in the first 3 months of treatment and is greater among women and elderly patients [ ].

Agranulocytosis after very long-term clozapine therapy has been reported [ ].

• A 41-year-old man suddenly developed agranulocytosis after taking clozapine nearly continuously for 89 months. During this time, his white blood cell and granulocyte counts remained stable. The white blood cell and granulocyte counts returned to baseline shortly after withdrawal of clozapine and administration of sargramostim.

Rechallenge : Cases of negative or positive rechallenge in patients with agranulocytosis have been reported [ , ].

• A 58-year-old man developed agranulocytosis during a second trial of clozapine, despite a successful previous trial [ ].

• A 17-year-old boy with severe clozapine-induced neutropenia had a negative rechallenge; because he had had an unsatisfactory response to traditional neuroleptic drugs, clozapine was continued despite a fall in white blood cell count, since concomitant treatment with granulocyte colony-stimulating factor was followed by rapid normalization of the white blood cell count [ ].

• A 29-year-old woman developed agranulocytosis after taking clozapine 300 mg/day for 5 years; 4 months after withdrawal, the clozapine was reintroduced (500 mg/day), and after 8 months the leukocyte count was still within the reference range [ ].

• A 45-year-old woman developed neutropenia after taking clozapine 500 mg/day for 6 years combined with other agents (olanzapine 10 mg/day, benzopril hydrochloride 20 mg/day, and haloperidol 150 mg/day) [ ]. Clozapine was withdrawn immediately and the granulocytes recovered within a few days. However, 10 weeks later clozapine was restarted and there was no recurrence over more than 3 years.

• A 40-year-old woman with clozapine-induced neutropenia prolonged by subsequent use of olanzapine was successfully re-challenged with both olanzapine and then clozapine [ ].

• A 35-year-old patient with resistant schizophrenia had three of episodes neutropenia while taking clozapine on separate occasions. Clozapine was later successfully reintroduced with filgrastim [ ].

Of 53 patients who were rechallenged with clozapine following leukopenia or neutropenia during previous clozapine therapy 20 had a further blood dyscrasia; in 17 of these 20 patients the second event was more severe, in 12 it lasted longer, and in 17 it occurred more quickly on rechallenge [ ]. Of the original 53 patients, 29 are still taking clozapine.

Prolongation of clozapine-induced leukopenia by olanzapine has been reported [ ].

Monitoring therapy : Over 10 000 patients have been treated with clozapine in Australia since its introduction in 1993, and the Clozaril monitoring system has ensured that since that time there have been no deaths from agranulocytosis in patients taking clozapine [ ].

An increase in white blood cell count of at least 15% above previous counts is a sensitive, although not specific, predictor for the development of agranulocytosis within 75 days [ ]. Clozapine dosage and baseline white cell count do not appear to predict agranulocytosis.

Monitoring G-CSF concentrations, if available, may be useful in following patients in whom clozapine-induced marrow damage is suspected.

An example of a false sense of security gained by relying on monitoring monthly blood counts in patients taking clozapine has been published [ ].

• A 61-year-old man who had taken clozapine for 3 years had normal blood counts. However, one day, his hemoglobin was 8.5 g/dl, having previously been 13 g/dl, following a steady asymptomatic fall over 6 months that had been documented but had gone unnoticed. He subsequently underwent investigation and treatment for anemia.

However, it is not clear in this case that clozapine was responsible for the anemia.

In a cohort study, based on a prospective drug exposure database, the effectiveness of centralized routine monitoring of blood counts was evaluated in 1500 patients taking clozapine between March 2001 and December 2001 [ ] Seven patients developed severe neutropenia while taking clozapine (neutrophil counts below 1.5 × 109/l). The mean time to withdrawal of therapy was 1.6 days (maximum 6 days), and neutrophil counts recovered to normal in all cases after 6.4 days (maximum 13 days). Based on an estimate of 500 patient-years of exposure, the frequency of severe neutropenia was one case per 71 patient-years of therapy or 1.4% per annum.

According to the recommended guidelines by Novartis, neutropenia (a white blood cell count below 3.0 × 109/l or an absolute neutrophil count below 1.5 × 109/l) during clozapine treatment is classified as being in the “red-alert zone”; immediate withdrawal of clozapine is recommended and reinstitution prohibited. However, in some patients, this is not feasible, because of lack of effective alternatives to clozapine. In five patients who were maintained on clozapine despite red-alert zone neutropenia and two control patients who discontinued clozapine because of neutropenia, hematological and clinical progress was followed for more than 600 days [ ]. In all five patients, there were no additional episodes of neutropenia despite continued clozapine treatment.

Treatment : Withdrawal of clozapine can lead to resolution of agranulocytosis, but not always. Granulocytopenia, presumably induced by clozapine, persisted in a 53-year-old woman after she switched from clozapine to quetiapine [ ].

Treatment with granulocyte colony-stimulating factor and granulocyte macrophage colony stimulating factors was helpful in a case of sepsis and neutropenia induced by clozapine [ ] and in a case of agranulocytosis in a 45-year-old man [ ].

Patients with leukopenia associated with clozapine have been successfully treated with lithium carbonate [ ], and another case has been reported in a 55-year-old man [ ].

Lithium can be used in combination with clozapine, and in these patients the possibility of inducing leukocytosis and increasing the total leukocyte count and the granulocyte count has been considered [ ]. Lithium has even been used to prevent clozapine-induced neutropenia [ ]. It has also been used in a patient with clozapine-induced neutropenia and in another with complete agranulocytosis: in both cases lithium increased the neutrophil count to within the reference range within 6 days [ ]. In the patient who had neutropenia, clozapine was restarted in the presence of lithium and the neutrophil count did not fall thereafter. Five other patients who took combined clozapine and lithium had a significant improvement with this combination and there were no cases of agranulocytosis, neuroleptic malignant syndrome, or other adverse effects [ ].