Clozapine-induced pneumonitis mimicking an atypical pneumonia

Abstract

Clozapine is an antipsychotic used for treatment-resistant schizophrenia with a significant side effect profile, including agranulocytosis, myocarditis and fever. Clozapine-induced fever often occurs in the first 2 weeks of treatment and settle after a few days. We report a case of a woman in her mid-30s who developed fever and infective symptoms suggestive of an atypical pneumonia while on clozapine titration. She was on clozapine for 16 days before developing high-grade fever, dry cough, diarrhoea, headache and photophobia with a very high CRP. We performed an extensive infection workup that returned negative results except for bilateral upper lobe ground glass changes of the lungs on CT. Despite antibiotic therapy, which would cover an atypical pneumonia, her CRP remained elevated and her fever persisted. Focus was directed to clozapine-induced pneumonitis as the cause for her symptoms. Her antibiotics were ceased, and clozapine was downtitrated. With the adjustment of her clozapine dose, her fevers and associated symptoms resolved, and CRP downtrended. Her fevers did not return when clozapine was uptitrated in the community subsequently. Clozapine-induced fever or other immune-allergic reactions should be systematically considered when patients develop fever during the initiation phase of clozapine therapy. Ruling out infective causes is desirable prior to attributing fevers to clozapine especially when they are accompanied by infective symptoms and high inflammatory markers. Careful downtitration of clozapine should be considered rather than abrupt cessation in managing clozapine-induced fevers and subsequent slow uptitration could be considered.

Background

Clozapine is a highly potent antipsychotic often used in treating patients with treatment-resistant schizophrenia.¹ Due to its impressive adverse effect profile, including potentially life-threatening complications such as agranulocytosis and myocarditis, clozapine is reserved as a second-line medication with careful dose titration.²

Clozapine-induced fever is an overlooked side effect of clozapine because of its relatively benign course. There is a wide variation in documented incident rates for clozapine-induced fever, ranging from 10% to 30%.^3–5 Most of the reported literature describes the onset of fever about 10–14 days after initiation of therapy, with average highest temperatures below 39°C and fever lasting for 3–5 days.^3–5 The incidence of fever is typically higher in patients who are elderly, have clozapine titrated by more than 50 mg/week or concurrently use valproate.³ Though the exact pathophysiology behind clozapine-induced fever is unclear, studies have suggested an interleukin-6-driven process.⁶

Varying treatment approaches exist for clozapine-induced fever. In some cohorts, clozapine use continued throughout the fever course with subsequent resolution of febrile events.⁵ Given the innocuous course of clozapine-induced fever, automatic discontinuation of clozapine at the onset of fever is discouraged, and instead, monitoring for other life-threatening side effects while continuing clozapine therapy is recommended. In cohorts where clozapine therapy was discontinued, most fevers responded to the cessation of clozapine for 48 hours, with subsequent reintroduction leading to fever recurrence in 14% of patients.³

Here, we report a case of a young woman on clozapine therapy presenting with high-grade fever, which mimicked an atypical pneumonia.

Case presentation

A young woman in her mid-30s was referred to the general medical team by the inpatient psychiatry team for fever of unknown origin. She had presented to the psychiatry unit 3 weeks before with psychosis on a background of treatment-resistant schizophrenia. She was commenced on clozapine 16 days prior at 12.5 mg and titrated to 50 mg in the morning, 150 mg at night, with a beneficial response for her psychotic symptoms. The day before her planned discharge date, she developed a fever up to 40°C with chills and rigours and was referred to general medicine for further evaluation. Alongside the fever, she felt generally unwell and experienced a dry cough, frontal headache, photophobia, lower back pain and watery diarrhoea. Her history did not reveal any other infective focus, and her examination was unremarkable except for photophobia, occasional scattered crepitations in lung fields and lower lumbar spine tenderness. Given the undifferentiated nature with ongoing high fevers, she received an extended panel of investigations with a broad differential to consider.

Investigations

Her initial white cell count was 8.3×10⁹/L and differential counts remained within normal limits throughout her course. Her c-reactive protien (CRP level) was 6.6 mg/L at the onset of the fever and increased to 283 mg/L over the next 6 days. The procalcitonin level measured on day 6 of fever was 4.93 µg/L. Serum biochemistry revealed a mild fluctuant transaminitis with aspartate transaminase (AST) and alanine transaminase (ALT) of 110–130 U/L. Her clozapine/norclozapine levels measured at the onset of fever were 540/230 µg/L, ratio 2.3, and on day 6 from fever onset were 930/390 µg/L, ratio 2.4 (250–600 µg/L). Troponin levels remained normal, at onset of fever 4 ng/L and repeat level on day 7 was 5 ng/L (34 ng/L). She provided and returned normal results for urine culture, blood culture, stool viral/bacterial/protozoal PCR, extended respiratory viral panel, pneumonia screen for mycoplasma/legionella/chlamydia/Streptococcus pneumoniae, HIV Ab, EBV IgM and CMV IgM. An MRI spine to exclude an epidural abscess and a lumbar puncture (LP) to exclude meningoencephalitis were completed and unremarkable. A transthoracic echo also excluded myopericarditis/infective endocarditis. Her chest X-ray was unremarkable. She received a CT pan scan to assess for an infective focus, which demonstrated bilateral upper peribronchovascular changes suggestive of an infective/inflammatory aetiology (figure 1).

Figure 1.

CT chest showing bilateral upper peribronchovascular changes suggestive of an infective/inflammatory aetiology. (A) In coronal view. (B) In transverse view.

Differential diagnosis

She carried a broad differential, given the undifferentiated nature of her presentation. These included infective, inflammatory, malignant and drug-induced aetiologies. Provided the symptoms, she demonstrated were suggestive of an infection with a very high CRP, an infective cause for her fever ranked prominently and required exclusion before denoting her fevers as a benign side effect of clozapine.

Treatment

She was empirically commenced on intravenous flucloxacillin 2 g 6 hourly and intravenous gentamicin 5 mg/kg daily for fever of unknown origin at presentation. This regimen was changed to a meningitic cover with intravenous ceftriaxone 2 g daily and aciclovir 10 mg/kg three times a day when her headache and photophobia became more prominent until LP results excluded meningitis. With a negative LP result and pan-CT scan revealing subtle chest signs, we changed her antibiotics to intravenous piperacillin–tazobactam 4.5 g 6 hourly and intravenous azithromycin 500 mg daily to cover for a possible atypical pneumonia.

Given that her psychotic symptoms improved with clozapine therapy, the psychiatry team wanted to continue her clozapine at the current dose, especially with normal clozapine levels at the onset of fever. However, with her infective and autoimmune screening returning normal results and fevers remaining persistently high despite antibiotic therapy, and high clozapine levels on day 6 of the fever, clozapine was considered the cause for her symptoms. Antibiotics were ceased, and clozapine downtitration was commenced on day 6 of her presentation with a reduction to 50 mg two times per day for 2 days and subsequently reduced to 25 mg two times per day. With the decrease of her clozapine dose, her febrile, constitutional and infective symptoms disappeared, and her CRP downtrended to 39 mg/L over the next 3 days.

Changes in her fever, inflammatory markers, clozapine levels and dose titration are summarised in figure 2.

Figure 2.

Change in clinical and biochemical parameters. Abbreviations: CRP (c-reactive protein)

Outcome and follow-up

Following the resolution of her fever, she was transferred back to the mental health ward for observation of her psychotic symptoms and discharged 4 days later. She was followed up weekly in the community mental health centre and had her clozapine dose titrated up weekly over the next month for residual psychotic symptoms to a dose of 50 mg in the morning and 125 mg at night without recurrence of fever or other clozapine associated adverse effects.

Discussion

Clozapine-induced fever is mostly noted during the initiation phase of therapy, whereas our patient had been on clozapine over 2 weeks before she developed fever. The literature reports mean peak temperatures of 39°C that typically last 3–4 days^3–5. Our patient had temperatures up to 40°C and continued to have high-grade fevers even on day 6 after the onset of fever.

Most patients with clozapine-induced fever have generalised constitutional symptoms with a benign transient course.⁷ Our patient demonstrated multiple infective symptoms, such as dry cough, severe headache, photophobia and diarrhoea. Though elevated CRP has been recorded in previous cases of clozapine-induced fever and associated complications, very high CRP as recorded in our patient, is not standard in other reported cases of clozapine-induced fever.^8–10 Furthermore, our patient had elevated procalcitonin, which is not reported in reviewed case reports. These indicators prompted us to investigate our patient in-depth for an infectious cause.

Clozapine is strongly associated with risk of pneumonia and associated mortality; it increases the risk of aspiration secondary to sedation and hypersalivation as well as other pneumonias secondary to its effect on immunity.¹⁰ Our patient had a normal chest X-ray but demonstrated subtle bilateral upper lobar changes on her CT which we initially treated as atypical pneumonia. Despite adequate antibiotic therapy which would cover aspiration or atypical pneumonia, our patient continued to have fevers.

Pneumonitis is another reported complication of clozapine therapy.¹¹ The bilateral upper peribronchovascular changes noted in our patient’s CT may reflect a similar process. Her cough resolved along with her fever on clozapine downtitration, making clozapine-induced pneumonitis the likely cause in our patient. The Naranjo scale returned a score of 9 in this patient’s case, based on the temporal relationship symptoms and objective evidence had with the titration of clozapine, its blood levels and having other causes excluded adequately, suggestive of a definite probability of an adverse drug reaction.¹² Given the risk–benefit of repeat CT, she did not receive a follow-up CT to establish if these bilateral upper peribronchovascular changes resolved.

The patient also had diarrhoea while febrile with a negative infective screen. Though constipation is the more common gastrointestinal complication of clozapine therapy, this medication is also reported to cause diarrhoea, especially in the initiation phase of therapy.¹³

Clozapine is largely metabolised by Cytochrome P450-1A2 isozyme. Inflammatory and infective processes can inhibit this isozyme via cytokine mediated pathways leading to increased levels of clozapine which in turn will bind to other acute phase proteins and subsequent variable manifestations of clinical toxicity. Our patient had high clozapine levels when measured on day 6 of the illness. During the inflammatory phase, her clozapine:norclozapine ratio remained high and downtrended as the inflammatory response settled. Other than the clozapine dose, these high clozapine levels along with high clozapine:norclozapine ratio could have been contributed to by the inflammatory response in her lungs, especially given she did not have other clinical manifestations of clozapine toxicity.¹⁴

Our patient’s fever, infective symptoms and CRP responded well to cautious downtitration of her clozapine. Though the literature suggests continuation of clozapine therapy while allowing for fever to settle, we opted for downtitration given she had ongoing symptoms even on day 6 of presentation.⁷ She required uptitration of her clozapine in the community to manage residual psychotic symptoms without the recurrence of fever.

Learning points.

• Clozapine-induced fever or other immune-allergic reactions should be systematically considered when patients develop fever during the initiation phase of clozapine therapy and should be considered early in the course of the illness irrespective of clozapine levels.

• Associated infective symptoms and very high inflammatory markers could mislead towards an infective cause of fever, but it would be prudent to exclude infection before attributing the fever to clozapine.

• Clinicians should be aware that inflammatory/infective processes can increase the clozapine levels with variable clinical manifestations, and therefore, the need for close monitoring of clozapine levels in patients with acute illness.

• In patients suspected to have clozapine-induced fever, careful downtitration should be considered instead of abrupt cessation if the psychotic symptoms are well controlled on clozapine.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).