Neonatal Perinatal Fellow University of Colorado School of Medicine Morrison, Colorado, United States
Background: Mechanically ventilated (MV) infants in the Neonatal Intensive Care Unit (NICU) are highly susceptible to ventilator-associated infections (VAI). To aid in the diagnosis of these infections, tracheal aspirate (TA) cultures are commonly used, and bacterial growth is often treated with antibiotics. Studies suggest that TA cultures have a low sensitivity and specificity to diagnose a VAI and may lead to potential overuse of antibiotics, which is associated with adverse outcomes. Objective: We aimed to investigate short term outcomes associated with antibiotic treatment for VAI. Design/Methods: Retrospective cohort study evaluating infants with and without antibiotic exposure for VAI’s. MV infants in a Level IV NICU who had a TA culture obtained in 2019. All TA cultures during their NICU stay were analyzed. Antibiotic exposure was ≥3 days of treatment for a VAI. Clinical variables and details of each TA sample were obtained from the electronic medical record. Results: 308 ETA cultures in 71 infants were analyzed; of these, 65% of infants received antibiotics for a VAI. Gestational age, birth weight, age in days, clinical diagnoses, and indications for TA culture collection were not significantly different between groups. Respiratory factors such as ventilator days, and BPD classification at 36 weeks corrected age were not significantly different between groups. Polymorphonuclear (PMN) cells on TA culture were more likely to be moderate to heavy (61%) in the antibiotic group, and none to few in the no antibiotic group (73%). Infants in the antibiotic group had more TA cultures sent (median 4 vs 1; p<.01), increased average charge associated to TA (median $1272 vs $318; p<.01), longer length of NICU stay (median 111 vs 79 days; p=.05), and were more likely to have tracheostomy placement (33% vs8%; p<.05). No difference between groups for discharge on supplemental oxygen or in-hospital death.
Conclusion(s): Most MV infants in our Level IV NICU had antibiotics administered for a suspected VAI. Baseline characteristics were similar in both groups, suggesting the decision to treat was not significantly influenced by underlying patient clinical characteristics. Antibiotic treatment was highly correlated with presence of PMNs on TA and was associated with higher number of TA cultures obtained and increased charges. Interestingly, infants receiving treatment for VAI were more likely to ultimately receive tracheostomy for long-term MV. While our study cannot demonstrate causality, further investigation is warranted to determine the impact of respiratory infections on evolving neonatal chronic lung disease.
