Pediatric Antifungal Utilization: New Drugs, New Trends

Priya A. Prasad, MPH; Susan E. Coffin, MD, MPH; Kateri H. Leckerman, MS; Thomas J. Walsh, MD; Theoklis E. Zaoutis, MD, MSCE


Pediatr Infect Dis J. 2008;27(12):1083-1088. 

In This Article


Since 2000, there have been significant changes in the prescribing of antifungal agents in hospitalized children. Although AMB has been considered the standard in antifungal therapy during the past 4 decades,[1,11,13,14] pediatricians have markedly decreased their utilization of this agent. The data in this study indicate that LFAB and the newer agents, echinocandins and voriconazole, replaced AMB after their introduction. Although AMB prescriptions also decreased significantly in neonates, it is still commonly used for the treatment of candidiasis. This observation is consistent with data demonstrating that infants and young children infrequently experience adverse events associated with AMB and LFAB;[14] in addition, there is a lack of dosing and safety data for the prescription of newer agents in this population. The increase in the number of neonates requiring antifungal therapy over the study period likely reflects the increased use of prophylactic fluconazole to prevent invasive fungal infections in premature neonates.[18,19]

As the antifungal treatment options available to clinicians have increased, prescribing and treatment strategies have also significantly changed. For example, studies have shown that the echinocandins are at least as effective as AMB for the treatment of candidiasis and their use is associated with fewer side-effects.[20] Although no direct comparisons have been made between AMB and the echinocandins within a large pediatric cohort, data from a prospective multicenter study show that the efficacy of the echinocandins in pediatric patients with invasive aspergillosis or invasive candidiasis is consistent with results from previous adult studies.[21] Our data show that later in the study period, clinicians were more commonly prescribing the echinocandins for disseminated/systemic candidiasis and other fungal infections than AMB and LFAB. Both caspofungin and micafungin have been studied in children; caspofungin administered at 50 mg/m2/d in children provides exposure that is comparable to that after 50 mg/d doses in adults.[22] Micafungin given in a dose of 4-5 mg/kg has been shown to be as effective as LFAB at treating invasive candidiasis in children.[23]

The most striking change in antifungal prescribing was seen in children with aspergillosis, as AMB was almost completely replaced by voriconazole. We assume that this change in practice was influenced by the large randomized clinical trial demonstrating that voriconazole is statistically superior to AMB for treatment of invasive aspergillosis. Notably, however, few children were included in this study.[24] Because of the more accelerated metabolic clearance in pediatric patients, the dosages of voriconazole may need to be higher.[25] A maintenance dose of 7 mg/kg twice daily in children is recommended by the European Medicines Agency for the attainment of plasma values that are comparable to those in adults. Loading regimens in pediatric populations have not been adequately studied.

The majority of patients prescribed an antifungal agent did not have a documented fungal infection by ICD-9-CM code. This finding may be due in part to the low sensitivity of ICD-9-CM codes for identifying fungal infections. However, we determined that 8% of these patients were diagnosed with septicemia and 5% were diagnosed with bacteremia, which could account for some of the antifungal treatment. In addition, patients who did not have a documented fungal infection were most commonly prescribed fluconazole (77%) and were more likely to have an underlying condition (82%) when compared with those patients who were diagnosed with a fungal infection (72%). Because of their underlying condition, these patients without documented fungal infection could have been receiving antifungal therapy as prophylaxis or empiric therapy.

Because of the comprehensive data on more than 6 million patients, the PHIS database can be a powerful tool for clinical researchers. This descriptive study included a large number of geographically diverse and representative freestanding children's hospitals, suggesting that the results could be generalizable to other regions of the country. Our study has several limitations to be considered. First, the determination of fungal infection status and/or underlying comorbid conditions may be vulnerable to miscoding as these variables were determined by ICD-9-CM codes; however, fungal codes tend to have high specificity. Therefore the results of our study would more likely represent an underestimate of the burden of disease in the population. Additionally, the CCCs used to designate underlying condition in this analysis were previously validated by Feudtner et al.[16] Although there are inherent limitations associated with the use of administrative data, we believe our results accurately describe national pediatric antifungal prescribing patterns.

In summary, we found that the number of hospitalized children prescribed antifungal therapy has increased and the choice of antifungal therapy has changed dramatically with the introduction of newer antifungal agents. This rapid shift to the use of newer agents has occurred despite the lack of adequate pharmacokinetic/pharmacodynamic, safety, or efficacy data in children. The shift likely reflects clinicians' desire to use agents with better toxicity profiles. Further studies must be conducted to determine the optimal dosing, efficacy, and safety of these newer agents in the pediatric population.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.