Antibiotic Overuse Among Children with COVID-19 Hospitalized in a Pediatric Intensive Care Unit

ABSTRACT

Background. Antibiotic overuse has been well described among hospitalized adults with COVID-19 but similar evaluations in children are lacking. We sought to quantify bacterial infection rates and antibiotic utilization among critically ill children hospitalized with COVID-19 to identify opportunities to optimize care. Methods. We performed a single center retrospective cohort study of all children hospitalized with symptomatic COVID-19 in a pediatric intensive care unit between May 16, 2020 and February 11, 2022 at a tertiary care children's hospital in the Southeastern U.S. We performed medical record review to demographic and clinical characteristics. This study was approved by the institutional IRB with a waiver of consent. Results. During the study period there were 92 subjects hospitalized in the intensive care unit with COVID-19. Demographic and clinical characteristics of the cohort are summarized in the Table. Median age was 12.4 years, median length of stay was 6 days, 32% of subjects required mechanical ventilation and 5% died. The vast majority of children had one or more comorbidities and only 1 subject was fully vaccinated against SARS-CoV-2. Thirteen (14%) subjects had bacterial growth from any clinical specimen. Eight subjects had respiratory cultures that may have represented airway colonization;when these were excluded, 5 (5%) subjects had either urinary tract or bloodstream infections. Two of the bloodstream infections were caused by drug-resistant organisms and were hospital-acquired. Despite the low number of subjects with bacterial infections, 45% received antibiotics for >3 days. The antibiotic days of therapy per subject varied widely and ranged from 0 to 61 days. Conclusion. In this cohort of nearly 100 critically ill children with COVID-19, the rate of culture-confirmed bacterial infection ranged from 5-14% yet nearly half of patients received antibiotics. Limitations include the single center and retrospective study design and the fact that bacterial pneumonia may not be cultureconfirmed. Despite these limitations, this work suggests that children with COVID-19 rarely have bacterial co-infections and are often prescribed unnecessary antibiotics.