Voriconazole drug monitoring in the management of invasive fungal infection in immunocompromised children: a prospective study.

Objectives: To evaluate voriconazole plasma level monitoring in immunocompromised children and determine the relationship of plasma levels with dose, safety and efficacy.

Methods: We used a prospective study including all consecutive children with invasive fungal infection (IFI) treated with voriconazole between August 2008 and May 2010. IFI diagnosis and clinical outcome evaluation were based on European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group ('EORTC/MSG') definitions.

Results: A total of 196 voriconazole plasma trough measurements from 30 patients (median age 10 years) obtained during 2135 days of voriconazole therapy were analysed. Nineteen patients (63%) presented with proven or probable IFI. Voriconazole plasma levels varied widely and 73% of patients required dose adjustment. The median voriconazole dose was 20 mg/kg/day and the median duration of therapy was 6 weeks. Age 5 was the smallest value defining two groups on which the correlation between dose and plasma levels had a different behaviour, and this relationship was especially significant for patients <5 years old (Spearman's rank correlation coefficient=0.38213, P=0.008). For patients <5 years old the median dose to achieve therapeutic levels was 38.0 mg/kg/day (12-40.0) and for those ≥5 years old it was 15 mg/kg (4-52). Voriconazole plasma levels showed a significant relationship with early outcome (P=0.0268), but not late outcome (P=0.2015). Overall mortality was 42% and a significant relationship with voriconazole therapeutic plasma levels was not demonstrated. A significant relationship was established between plasma levels above normal range and skin and neurological toxicity (P=0.0001), but this could not be demonstrated for liver toxicity.

Conclusions: Our study confirms the large variability in voriconazole trough plasma levels in children and a trend to non-linear pharmacokinetics in older patients. In addition, doses significantly higher than those recommended in younger children seem warranted and a significant relationship between plasma voriconazole above the normal range and some adverse events is confirmed.