Ellsworth, Kevin

Introduction: Invasive candidiasis has a high morbidity and mortality among premature neonates. Antifungal prophylaxis with fluconazole significantly lowers the risk of invasive fungal infection in this population. We noted the use of fluconazole prophylaxis in our level IV neonatal intensive care unit (NICU) was variable and sought to standardize prescribing of prophylactic fluconazole. Methods: We formed a multidisciplinary team to develop an evidence-based protocol using literature and expert consensus to guide appropriate use of fluconazole prophylaxis in our level IV NICU. After determining baseline fluconazole prophylaxis prescribing before protocol implementation, we used plan-do-study-act (PDSA) cycles to introduce protocolized prescribing and incorporate it into daily practice. A 6-month intervention phase was followed by a 2-year control phase, in which monthly audits were performed to evaluate protocol adherence. Results were displayed in a statistical process control chart. Results: Before protocol implementation, fluconazole prophylaxis prescribing adhered to the protocol in 81% of patients. During the first PDSA cycle, adherence increased significantly to 94.5% (86/91 patients), which further increased to 98.7% (74/75 patients) during the second PDSA cycle and remained at 96% (120/125 patients) during the control phase (P < 0.0001). Conclusions: A multidisciplinary group-designed protocol was successful in standardizing fluconazole prophylaxis prescribing for infants in the level IV NICU. Adherence to protocol was high following implementation and was sustained for the duration of the project. There were no cases of invasive candidiasis noted.

Objective Among neonates of 22 to 29 weeks' gestational age (GA) who required mechanical ventilation for the treatment of respiratory distress syndrome (RDS) and clinically diagnosed pulmonary hypertension (PH), we tested our hypothesis that the association between early treatment with inhaled nitric oxide (iNO) and survival would vary according to birth size and GA. Study Design Because iNO was not randomly prescribed to patients in this cohort, we used propensity score matching to pair a neonate who received iNO at a chronological age of ≤7 days with an unexposed neonate with similar baseline characteristics. The primary outcome was inhospital mortality, which we evaluated based on size for GA and GA strata using the Cox proportional hazards regression model. Results Among 1,531 neonates who met study criteria, we created a propensity score matched cohort of 615 pairs of neonates (iNO-exposed and unexposed). The risk of inhospital mortality for iNO-exposed neonates was observed only in the minority (<10%) who were large for GA, though this finding did not persist when matching for illness severity. Conclusion Early treatment with iNO is not associated with survival in most extremely premature neonates with RDS and clinically diagnosed PH when stratified for birth size or GA.