Discussion

Although this particular interaction is discussed in a number of reviews, ^{Reference Chetty and Murray8,Reference Urichik, Prior, Dursun and Baker9,Reference Finch, Chrisman, Baciewicz and Self12,Reference Baciewicz, Chrisman, Finch and Seif15,Reference Prior, Chue, Tibbo and Baker21–Reference Spina, Scordo and D'Arrigo24} it is predominantly based on one case report from 1998 ^{Reference Joos, Udo and Wolfgang25} and a few refer to a second case published nearly a decade later. ^{Reference Peritogiannis, Pappas, Antoniou, Hyphantis and Mavreas26} In the first case, ^{Reference Joos, Udo and Wolfgang25} a 33-year-old male with schizophrenia who had a very good response to clozapine developed tuberculosis and was treated with rifampicin 600 mg daily, isoniazid and pyrazinamide. Within 2–3 weeks his clozapine serum levels decreased approximately sixfold and were subtherapeutic, and at 3.5 weeks psychotic symptoms returned. Prescribers increased his clozapine from 400 mg to 600 mg/day, but this did not lead to any improvement. Rifampicin was then substituted with ciprofloxacin, and the clozapine reduced back to 400 mg/day. Three days later clozapine serum levels returned to therapeutic values and the patient's psychotic symptoms improved. The second case ^{Reference Peritogiannis, Pappas, Antoniou, Hyphantis and Mavreas26} was a man with schizophrenia who was stable and well on clozapine 300 mg/day; rifampicin 600 mg/day was started for tuberculosis. Two weeks later his psychotic symptoms worsened and his previous clozapine dose-related adverse effects ceased. His clozapine was increased to 550 mg/day with only mild improvement. Six months later the rifampicin was stopped and his psychotic symptoms improved significantly and the previous dose-related adverse effects re-emerged. No clozapine levels were reported. In both cases the patient's previous clozapine dose-related adverse effects ceased when rifampicin was administered and their psychotic symptoms returned. Increasing the clozapine dose was not sufficient to manage the interaction. ^{Reference Baxter and Preston18}

In another case ^{Reference Panesar, Pai and Valachova27} where this combination was prescribed concurrently for 9 months the pharmacokinetic interaction was not mentioned, and although the patient did relapse, the main focus was the fact that they developed late-onset neutropenia following the introduction of rifampicin. Prior to that he had been mentally and haematologically stable on clozapine for 9 years.

Given these limited reports of this pharmacokinetic interaction, the description in standard texts is based on the theoretical knowledge of both drugs. This article describes a third case demonstrating the significance of this interaction, where clozapine was rendered ineffective by the potent enzyme induction of rifampicin. The serious psychotic relapse that ensued led to the patient's first psychiatric admission in about 8 years.

Generally, enzyme induction interactions are slow to develop and slow to resolve. Induction may take several days or 2–3 weeks to fully develop; ^{Reference Niemi, Backman, Fromm, Neuvonen and Kivisti13,Reference Virani, Bezchlibnyk-Butler and Jeffries17} the extent of enzyme induction depends on both the drug and its dose. ^{Reference Reitman, Chu, Cai, Yabut, Venkatasubramanian and Zajic14} The induction may persist for a similar period after stopping the inducer, although this varies considerably between drugs. ^{Reference Reitman, Chu, Cai, Yabut, Venkatasubramanian and Zajic14,Reference Baxter and Preston18} Rifampicin's onset effect on clozapine is thought to be 2–3 weeks, ^{Reference Baxter and Preston18,Reference Joos, Udo and Wolfgang25} and induction is near maximal at 300 mg/day. ^{Reference Niemi, Backman, Fromm, Neuvonen and Kivisti13}

The case described here was in keeping with these facts. The impact of the interaction on the patient's presentation was clinically significant at about 3 months. Although rifampicin's effect on the enzymes was likely to have lasted for at least 1–2 weeks after the drug was discontinued, ^{Reference Finch, Chrisman, Baciewicz and Self12,Reference Baxter and Preston18} the effect on the patient's clinical presentation took much longer to resolve. It could have been anticipated that the patient's clozapine maintenance dose (100 mg daily) would not lead to therapeutic serum levels but unfortunately none were recorded in that or the preceding year. Following the introduction and cessation of rifampicin, the patient's dose of clozapine was increased approximately sixfold, suggesting that the enzyme induction was very potent. From an interaction perspective, an alternative antipsychotic that is not induced by rifampicin ^{Reference Baxter and Preston18} could have been considered, rather than persisting with clozapine in the face of this interaction. However, this option was not pursued as the patient's schizophrenia was treatment resistant; several alternative antipsychotics had previously been tried but the patient failed to respond to them. Therefore, the decision was to optimise the clozapine dose carefully using therapeutic levels as a guide, aware that on withdrawal of an enzyme inducer plasma concentrations of the induced drug may increase and toxicity may occur. ¹¹

In the Summary of Product Characteristics (SPC) for rifampicin the only antipsychotics identified as interacting with it are haloperidol and aripiprazole, ¹⁰ although lurasidone and clozapine are also mentioned in the BNF. ¹¹

The incidence of tuberculosis in England had increased steadily from the late 1980s to 2005, and has remained at relatively high levels ever since. England has one of the highest tuberculosis rates in Western Europe. The incidence is more than four times higher than in the USA. If current trends continue, England will have more tuberculosis cases than the whole of the USA within 2 years. ²⁸ Therefore it is anticipated that there will be more patients in whom both clozapine and rifampicin are indicated, and this will need to be carefully managed.