Respiratory Infections Rather Than Antibiotics May Increase Clozapine Levels: A Critical Review of the Literature

Sir: Approximately 70% of clozapine metabolism is explained by the cytochrome P450 1A2 (CYP1A2). The CYP1A2 can be inhibited by respiratory infections. Decreased theophylline (a CYP1A2 substrate) clearance during upper respiratory infections has been described in asthmatic children and in adults with chronic bronchitis. Cytokines released during infection decrease CYP1A2 activity and synthesis. Expert pediatricians recommend decreasing theophylline doses by half during severe respiratory infections. Two cases of increased plasma clozapine concentrations during respiratory infection were recently published. The English-language literature was reviewed in October 2003 with a PubMed search using all searchable years and several term combinations including clozapine and CYP3A, clozapine and macrolide, and clozapine and fluoroquinolone. All the references listed in the newly found articles and in those gathered over the course of my more than 7 years of interest in the subject were also reviewed to obtain all relevant articles. This critical review of the literature suggests that 3 prior clozapine toxicity cases attributed to drug interactions with erythromycin or ampicillin are better explained by 1 pharyngitis case and 2 sinusitis cases. Ampicillin does not inhibit the CYPs. Erythromycin inhibits CYP3A, but a study of a single clozapine dose in 12 healthy males failed to show any effects of erythromycin. Similarly, itraconazole, another potent CYP3A inhibitor, did not influence clozapine metabolism. Therefore, in 2 patients taking erythromycin, CYP1A2 inhibition associated with a respiratory infection was probably the major factor contributing to clozapine toxicity, although some small effects of erythromycin cannot be ruled out. Assuming that clozapine follows linear pharmacokinetics, clozapine concentration-todose ratio associated with the respiratory infection decreased by a factor of 2 to 3 times (3.3, 2.0, 1.9, and 3.0). Clinicians caring for adult patients taking clozapine must therefore be careful should a patient develop serious respiratory infection with fever; they must pay particular attention to any signs suggesting clozapine toxicity (severe sedation, myoclonus, or even seizures). If any of these signs appear, the physician may need to decrease the clozapine dose, at least by half, until the patient has recovered from the infection. Obviously, this recommendation is limited by the small amount of published information supporting it (4 case reports); however, prospective studies in patients who are taking clozapine and having serious respiratory infections are unlikely to happen due to ethical and practical issues. The aim of this letter is to raise clinician awareness and stimulate the publication of new case report data that may or may not support the recommendation of decreasing the dose by half in clozapine-treated patients suffering a serious respiratory infection and showing signs of clozapine toxicity. CYP1A2 also metabolizes olanzapine, an antipsychotic agent used more frequently than clozapine. The limited experience my colleague and I reported in 1 patient taking both clozapine and olanzapine suggests that olanzapine levels increased by a factor of 1.7 during severe respiratory infection. An increase by a factor of 2 (or 1.7) in olanzapine level may not have the same clinical implications as in clozapine levels; olanzapine has a much wider therapeutic window. Finally, 3 clarifications are needed. (1) Not all antibiotics are free of drug interactions with clozapine; ciprofloxacin, a CYP1A2 inhibitor, can increase clozapine levels. (2) Smoking withdrawal for a few days during a respiratory infection may not have major effects on clozapine levels; the published case reports of clozapine toxicity after smoking cessation suggest that the increase of clozapine levels after smoking cessation usually takes at least 2 to 4 weeks to manifest. (3) Severe respiratory infections in patients taking clozapine can be associated with clozapine-induced agranulocytosis. According to data from the Clozaril National Registry, recent U.S. estimates suggest that agranulocytosis is less frequent (0.6%) than previously thought. Moreover, clozapine-induced agranulocytosis is very rare after 18 months of treatment and can be easily ruled out by drawing a white blood cell count.