Objective: Ventilator-associated pneumonia is among the most common nosocomial infections in the PICU. Respiratory secretion cultures and Gram stains are frequently obtained for diagnosis and to guide therapy, but their specificity is questionable. We conducted a scenario-based survey of pediatric intensivists to assess their antibiotic use in response to hypothetical tracheal aspirate culture and Gram stain results.
Design: Scenario-based survey.
Setting: A hypothetical PICU.
Patients: Three hypothetical scenarios of intubated children with fever and leukocytosis: a 4-month-old child with respiratory syncytial virus infection; a 7-year-old child with acute respiratory distress syndrome; and a 10-year-old child with aspiration pneumonia.
Interventions: Scenario-based survey of pediatric intensivists from the Pediatric Acute Lung Injury and Sepsis Network.
Measurements and main results: Ninety-four percent of the pediatric intensivists surveyed would obtain a respiratory secretion culture and Gram stain in the evaluation of an intubated child with fever and leukocytosis, most by simple tracheal aspiration but a minority (32%) by bronchoalveolar lavage. "Bacterial pathogenicity" was considered the most important result of the analysis. Although there were some differences across the three scenarios, most would initiate antibiotics if culture results identified methicillin-sensitive or methicillin-resistant Staphylococcus aureus or Pseudomonas and, on average, continue antibiotics for 7-10 days.
Conclusions: The majority of pediatric intensivists would obtain respiratory secretion cultures and Gram stains in the evaluation of an intubated child with fever and leukocytosis and initiate antibiotics guided by the results. The specificity of respiratory secretion cultures and Gram stains for the diagnosis of ventilator-associated pneumonia requires critical evaluation as this diagnosis is responsible for more than half of antibiotic use in the PICU.