This article is more than 5 years old. Information may no longer be current.
Decision tree promotes correct antibiotic prescribing for bloodstream infections
Click Here to Manage Email Alerts
Findings published in the Journal of the Pediatric Infectious Diseases Society suggest that a decision tree may be helpful in determining which children with bloodstream infections are at increased risk for developing resistance to broad-spectrum antibiotics.
Anna C. Sick-Samuels, MD, a pediatric infectious disease fellow at Johns Hopkins School of Medicine, told Infectious Diseases in Children that a clinical decision support tool has the potential to optimize empiric antibiotic selection for gram-negative bloodstream infections before antibiotic susceptibility testing is available. It can help distinguish children at risk for antibiotic resistance who would benefit from broader empirical therapy from patients for whom standard antibiotic treatment may be appropriate.
“Both aspects are valuable as time to appropriate antibiotic treatment is crucial for critically ill patients, but exposure to broad-spectrum antibiotic agents may also increase the risk for future resistant infections,” she said.
Sick-Samuels and colleagues conducted a longitudinal retrospective study that included children treated at a tertiary-care pediatric hospital between June 2009 and June 2015. During the study period, 689 gram-negative bloodstream infections occurred. Nearly one-third of these infections (31%) were resistant to broad-spectrum antibiotic treatment.
The researchers implemented the decision tree, which used five clinical characteristics to determine which children were at high or low risk of resistance, including:
- previous carbapenem treatment;
- a culture with a broad-spectrum antibiotic-resistant, gram-negative organism in the past 6 months;
- intestinal transplantation;
- age 3 years or older; and
- seven or more prior episodes of gram-negative bloodstream infections.
Sick-Samuels and colleagues observed that the decision tree could classify children with high risk for antibiotic-resistant infections with 46% sensitivity and 91% specificity.
The researchers acknowledged the tool’s low sensitivity, suggesting that there might be a risk for undertreating or overtreating children who are misclassified as being at high or low risk for developing resistance to broad-spectrum antibiotics.
“We are excited about the prospects for such a tool to support clinical decision making in pediatrics. However, this model should be validated in additional populations before it is ready to be implemented in practice,” Sick-Samuels said. “Additionally, it would be important to examine how this model would change clinicians’ antibiotic selection compared to decisions made without a prediction tool and if changes in antibiotic selection would have an impact on patient outcomes. Though additional studies are needed, this project reinforces the importance of considering a patient’s microbiology history and antibiotic exposure when selecting empiric antibiotics.” – by Katherine Bortz
Disclosures: Sick-Samuels reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.
Click Here to Manage Email Alerts