Discussion

Our survey revealed some of the antifungal stewardship practices and perceptions in a cohort of pediatric ASPs. Most pediatric ASPs have implemented some form of antifungal stewardship, and the strategies employed are highly variable across surveyed hospitals. Interestingly, although nearly half of respondents perceived inappropriate antifungal use as a problem at their hospital, many did not feel confident in providing antifungal recommendations. Specific stewardship strategies, including antifungal TDM and fungal susceptibility reporting, appear to be less common and may represent unique and important antifungal stewardship opportunities.

According to our survey results, the core antimicrobial stewardship strategies, PAF and preauthorization, have been expanded to include antifungals in most pediatric ASPs. Although these antifungal stewardship interventions have been shown to successfully reduce antifungal consumption in hospitalized patients, the best approach to antifungal stewardship remains unknown. ^{Reference Santiago-Garcia, Rincon-Lopez and Ponce Salas5,Reference Menichetti, Bertolino and Sozio17–Reference Valerio, Munoz and Rodriguez20} This lack of comparative data has likely led to the very heterogenous and hospital-specific approaches illustrated in this study. Some studies have proposed that PAF and more intensive stewardship strategies like daily PAF in combination with preauthorization and formulary restriction, may be more effective in decreasing antibiotic utilization. ^{Reference Stenehjem, Hersh and Buckel21,Reference Tamma, Avdic and Keenan22} Whether the same is true for antifungals is unknown. In our study, the use of preauthorization was more common than PAF, which is considered relatively labor intensive. ^{Reference Barlam, Cosgrove and Abbo14} Similar to prior reports evaluating antibiotic stewardship strategies, respondents reported that the biggest barrier to implementing antifungal PAF was a lack of time and resources. ^{Reference Hersh, Beekmann, Polgreen, Zaoutis and Newland23–Reference Newland, Gerber and Weissman25} The frequency of antifungal monitoring varied widely, with <10% of programs performing daily PAF. Whether the absence of daily PAF reflects a lack of sufficient resourcing or is a purposeful choice is uncertain; however, it may result in missed opportunities to intervene and improve antifungal prescribing. These findings highlight the importance of identifying effective antifungal stewardship interventions to better utilize and optimize current resources.

The characteristics of antifungal PAF and preauthorization varied across hospitals. For example, the survey detected differences in the types of antifungals monitored as part of PAF and restricted as part of preauthorization programs. Broad-spectrum triazoles, including voriconazole, posaconazole, and isavuconazole, were commonly targeted as part of antifungal stewardship efforts. ASPs may focus their efforts on these antifungals given the need for TDM, potential toxicities, and their higher costs. ^{Reference Munoz and Bouza26,Reference Ross, Hersh, Kronman, Newland and Gerber27} However, fluconazole may represent an important antifungal stewardship target because it accounts for ~70% of antifungal prescriptions in children ^{Reference Downes, Ellis, Lavigne, Bryan, Zaoutis and Fisher4} and has been associated with inappropriate prescribing in neonates and children. ^{Reference Lestner, Versporten and Doerholt1,Reference Ferreras-Antolin, Irwin and Atra28}

The TDM of azole medications is an important strategy to ensure the appropriate use of antifungals and mitigate adverse events. ^{Reference Park, Kim and Kim29} Most institutions responding to this survey routinely monitor azole levels; however, TDM was the least commonly monitored parameter among ASPs performing PAF of antifungals, and only a few institutions reported having a pharmacokinetics program to assist with antifungal TDM. These findings suggest that most clinicians conduct TDM of antifungals without expert guidance from pharmacists or ASP team members, which may increase the risk for medication errors and suboptimal dosing. Antifungals can be particularly challenging to use in children given their unique pharmacokinetic and pharmacodynamic characteristics. For example, studies have shown that up to 50% of antifungals are inadequately dosed and that certain antifungals require multiple dose adjustments to achieve therapeutic levels. ^{Reference Santiago-Garcia, Rincon-Lopez and Ponce Salas5,Reference Ferreras-Antolin, Irwin and Atra28,Reference Lachenmayr, Berking, Horns, Strobach, Ostermann and Berger30,Reference Soler-Palacin, Frick and Martin-Nalda31} Our study revealed variability in the target therapeutic levels reported across institutions, which suggests a lack of consensus in current practices and an important opportunity for standardization.

Many of the survey respondents reported a lack of confidence in making antifungal recommendations. This important finding could be related to the limited diagnostic information available to guide antifungal stewardship recommendations. Indeed, almost 40% of the respondents in our survey reported that having better diagnostics would improve their level of confidence in making antifungal recommendations. The absence of microbiological data may limit the ability of ASPs to identify opportunities for discontinuation or de-escalation of antifungals and contribute to prolonged courses of prophylactic or empiric antifungals. Previous surveys have shown that de-escalation of broad-spectrum antibiotic therapy on the basis of culture results is one of the most common antibiotic stewardship interventions. ^{Reference Hersh, Beekmann, Polgreen, Zaoutis and Newland23} In contrast, our study found that the recommendation to switch antifungals (ie, de-escalate therapy) was one of the least commonly reported interventions. As a surrogate for culture data, the use of noninvasive fungal markers was common among the surveyed hospitals. Prior studies have evaluated the use of noninvasive fungal markers for the surveillance of invasive fungal infection in stem cell transplant patients in an effort to shift from a prophylactic to a preemptive approach and decrease potentially unnecessary antifungal exposure. ^{Reference Cordonnier, Pautas and Maury32–Reference Morrissey, Chen and Sorrell34} Further studies exploring the safety and utility of these tests as part of antifungal stewardship interventions in the pediatric population are warranted.

Inappropriate and unnecessary antifungal use was considered a problem by most pediatric ASPs in this study. Oncology and stem cell transplant were reported as the services that would benefit the most from antifungal stewardship, likely because of the disproportionally higher antifungal utilization described in this population. ^{Reference Goldman, Ross and Lee7} Also, most ASPs in this cohort identified antifungal resistance as a national problem, and to a lesser extent, an institutional problem. This perception may be due to the recent emergence of multidrug-resistant Candida auris outbreaks in hospitals across the United States. ^{Reference Forsberg, Woodworth and Walters35} Most of these cases have been reported in adults; therefore, respondents of this study may consider issues of antifungal resistance as a national, rather than local, problem. Only 20% of the hospitals reported antifungal susceptibilities as part of the institutional antibiogram, and ASPs in these hospitals were more likely to perceive antifungal resistance as a problem at their institution. Possibly, therefore, resistant fungal organisms are underreported or ASP team members are not aware of institutional antifungal susceptibility trends. The emergence of antifungal resistance has been increasingly recognized, and understanding the local epidemiology is essential to developing institutional guidelines for antifungal prophylaxis and the empiric treatment of fungal infections.

Our study has several limitations. First, these results may not be generalizable to all pediatric hospitals. Although our response rate was high, we only surveyed ASPs participating in the SHARPS Collaborative, in which members are actively involved in identifying best practices for the use of antimicrobials. Therefore, we may be overestimating the extent to which antifungal stewardship practices are occurring. However, we were able to capture responses from both freestanding children’s hospitals and children’s hospitals within a larger adult hospital. In addition, ASPs from all 4 geographic regions in the United States and 2 international sites in Europe were represented in our cohort. Another limitation is that some of the responses may have been influenced by the respondent’s familiarity with their institution’s resources and practices and may represent perception rather than fact. Finally, we did not evaluate the effectiveness of different antifungal stewardship interventions, which should be further explored in future studies.

Although implementing antifungal stewardship has challenges, it has the potential to decrease unnecessary and suboptimal antifungal use, to reduce toxicities associated with inappropriate antifungal use, and to prevent the emergence of antifungal resistance. Currently, PAF and preauthorization are the main strategies used by pediatric ASPs to perform antifungal stewardship. Future studies should evaluate the effectiveness and clinical impact of such interventions. Optimization of antifungal dosing and TDM via ASPs represent key components of antifungal stewardship in pediatrics. Thus, future antifungal stewardship efforts should focus not only on reducing unnecessary use but also on improving the quality of antifungal prescribing.