Increasing the dose of voriconazole compensates for enzyme induction by phenytoin

We read with interest the manuscript by Dr Alffenaar, in which it was stated that doubling the dose of voriconazole does not compensate for induction by phenytoin [1]. Voriconazole, used as first-line agent to treat invasive aspergillosis in immunocompromised patients, is known to have complex pharmacokinetics resulting in variable plasma concentrations [2]. It is recommended to monitor voriconazole plasma concentrations in non-responsive patients, gastro-intestinal dysfunction, severe hepatopathy, unexplained neurological symptoms and in children. Target concentrations should be higher than 1 or 2 mg l⁻¹ to guarantee efficacy and lower than 6 mg l⁻¹ to avoid toxicity [2]. Also, co-administration of other drugs can result in fluctuating plasma concentrations as voriconazole is metabolized by CYP2C9, CYP2C19 and to a lesser extent by CYP3A4. Plasma concentrations are drastically lowered by CYP450 inducers including rifampicin, phenobarbital and carbamazepine. Consequently, concomitant use of these drugs is contra-indicated [3].

When phenytoin is co-administered, the peak concentration and AUC of voriconazole decrease by 49 and 69%, respectively, as shown in healthy volunteers [4]. Increasing the dose of voriconazole from 200 mg to 500 mg twice daily in the case of oral treatment, or from 4 mg kg⁻¹ to 5 mg kg⁻¹ twice daily in the case of i.v. administration, has been suggested to compensate for this effect [4, 5].

Recently, a patient was concomitantly treated with both phenytoin and voriconazole in our hospital. Trough concentrations were quantified by a validated HPLC analysis followed by UV detection. The patient was a 28 year old, woman weighing 60 kg with a history of kidney transplantation, who was admitted to the ICU because of respiratory failure secondary to probable invasive aspergillosis (cavitating lesion in the lung and serum galactomannan OI 5.0, nl <0.5). She was treated with voriconazole orally (200 mg twice daily after a loading dose). On day 4, she suffered from several epileptic insults, probably due to cerebral aspergillosis, for which phenytoin (loading dose of 1200 mg, followed by 100 mg three times daily) was started. As recommended [5], the maintenance dose of voriconazole was increased up to 5 mg kg⁻¹ twice daily (i.e. 300 mg twice daily) after 5 days of co-administration with phenytoin. Subsequently, voriconazole was switched to intravenous administration as her renal function was improving. Phenytoin was replaced by levetiracetam on day 12, and it was decided to reduce voriconazole dosage to 4 mg kg⁻¹ on day 22, i.e. 10 days after withdrawal of phenytoin.

Voriconazole trough concentrations were measured on several occasions, as shown in Figure 1. On days 9–11, during concomitant treatment with voriconazole and phenytoin, the mean plasma concentration was 2.6 mg l⁻¹. Concentrations were also determined on days 18 and 19, when phenytoin was withdrawn during which the patient was still treated with voriconazole at a dose of 5 mg kg⁻¹ twice daily; a mean plasma concentration of 2.5 mg l⁻¹ was measured. On days 23–25, after reduction of the dosage of voriconazole, a mean concentration of 3.5 mg l⁻¹ was found. The patient received no other medication interacting with voriconazole metabolism and liver function tests were normal.

Figure 1.

Chronology of voriconazole (VCR) plasma concentrations

Treating the patient with 5 mg kg⁻¹ i.v. seemed to counterbalance induction by phenytoin as trough concentrations stayed well within the therapeutic range of 2–6 mg l⁻¹ during and after phenytoin treatment. Our findings are in line with the results found by Purkins et al. and in contrast to the cases discussed by Allfenaar et al. [1, 4]. In the latter cases, concentrations were very low (0.1–0.2 mg l⁻¹) during co-treatment [1]. However, it was not mentioned how voriconazole was administered. In our opinion, voriconazole should preferably be administered intravenously to avoid a low bio-availability, especially in severely ill patients and in patients treated with potent inducers [2].

Very recently, it was shown that voriconazole exposure is three times higher in CYP2C19 poor metabolizers compared with extensive metabolizers [6]. In our patient, CYP2C19 genotype was not determined. However, as phenytoin steady-state concentrations were within normal ranges (mean 16.4 mg l⁻¹) and phenytoin is a CYP2C19 substrate, it was not expected that the patient was a CYP2C19 poor metabolizer.

The process of induction requires new CYP450 protein synthesis and may take days or even weeks to develop fully. Likewise, the effect may persist for a similar length of time following withdrawal of the inducer, dependent on the half-life of the inducer and the enzyme induced [7]. Raising the dose of the voriconazole may counteract the interaction, but this requires regular monitoring, as it is not clear when exactly the dose adjustment should happen. In our patient, concentrations measured 7 days after withdrawal of phenytoin, were still within the therapeutic range. It seems that reducing voriconazole on day 10 after phenytoin withdrawal is in line with good clinical practice.

In our opinion, voriconazole can be combined with phenytoin. However, the dose of voriconazole should be adequately increased. If possible, intravenous administration is preferred in critically ill patients. Concentrations should be determined on a regular basis during phenytoin treatment and after phenytoin withdrawal, as it is not clear when doses should be adjusted.

Competing interests

There are no competing interests to declare.