A study found that only 2% of children hospitalized with community-acquired pneumonia (CAP) actually had any causative pathogen in their blood culture results, despite national guidelines that recommend blood cultures for all children hospitalized with moderate to severe CAP.
The guidelines are the 2011 guidelines for managing CAP published by the Pediatric Infectious Diseases Society (PIDS) and the Infectious Diseases Society of America (IDSA) (Clin Infect Dis. 2011 Oct;53:617-30).
Cristin O. Fritz, MD, of the Children’s Hospital of Colorado, Aurora, and associates conducted a data analysis of the EPIC (Etiology of Pneumonia in the Community) study to estimate prevalence, risk factors, and clinical outcomes in children hospitalized with bacteremic CAP and to evaluate the relationship between positive blood culture results, empirical antibiotics, and changes in antibiotic treatment regimens.
Data were collected at two Tennessee hospitals and one Utah hospital during Jan. 1, 2010–June 30, 2012. Of the 2,358 children with CAP enrolled in the study, 2,143 (91%) with blood cultures were included in Dr. Fritz’s analysis. Of the 53 patients presenting with positive blood culture results, 46 (2%; 95% confidence interval: 1.6%-2.9%) were identified as having bacteremia. Half of all cases observed were caused by Streptococcus pneumoniae, with Staphylococcus aureus and Streptococcus pyogenes noted less frequently, according to the study published in Pediatrics.
A previous meta-analysis of smaller studies also found that children with CAP rarely had positive blood culture results, a pooled prevalence of 5% (Pediatr Infect Dis J. 2013;32:736-40). Although it is believed that positive blood culture results are key to narrowing the choice of antibiotic and predicting treatment outcomes, the literature – to date – reveals a paucity of data supporting this assumption.
Overall, children in the study presenting with bacteremia experienced more severe clinical outcomes, including longer length of stay, greater likelihood of ICU admission, and invasive mechanical ventilation and/or shock. The authors also observed that bacteremia was less likely to be detected in children given antibiotics after admission but before cultures were obtained (0.8% vs 3%; P = .021). Pleural effusion detected with chest radiograph also consistently indicated bacteremic pneumonia, an observation made within this and other similar studies.
Also of note in detection is the biomarker procalcitonin, which is typically present with bacterial disease. Dr. Fritz and colleagues stressed that because the procalcitonin rate was higher in patients presenting with bacteremia, “this information could influence decisions around culturing if results are rapidly available.” Risk-stratification tools also might serve a valuable purpose in ferreting out those patients presenting with moderate to severe pneumonia most at increased risk for bacterial CAP.
Compared with other studies reporting prevalence ranges of 1%-7%, the prevalence of bacteremia in this study is lower at 2%. The authors attributed the difference to a possible potential limitation with the other studies, for which culture data was only available for a median 47% of enrollees. Dr. Fritz and her colleagues caution that “because cultures were obtained at the discretion of the treating clinician in a majority of studies, blood cultures were likely obtained more often in those with more severe illness or who had not already received antibiotics.” In this scenario, the likelihood that prevalence of bacteremia was overestimated is noteworthy.
The authors observed that penicillin-susceptible S. pneumonia was the most common cause of bacteremic CAP. They further acknowledged that their study and findings by Neuman et al. in 2017 give credence to the joint 2011 PIDS/IDSA guideline recommending narrow-spectrum aminopenicillins specifically to treat children hospitalized due to suspected bacterial CAP.
Despite its small sample size, the results of this study clearly demonstrate that children with bacteremia because of S. pyogenes or S. aureus experience increased morbidity, compared with children with S. pneumoniae, they said
While this is acknowledged to be one of the largest studies of its kind to date, a key limitation was the small number of observable patients with bacteremia, which prevented the researchers from conducting a more in-depth analysis of risk factors and pathogen-specific differences. That one-fourth of patients received in-patient antibiotics before cultures could be collected also likely led to an underestimation of risk factors and misclassification bias. Lastly, the use of blood culture instead of whole-blood polymerase chain reaction, which is known to be more sensitive, also may have led to underestimation of overall bacteremia prevalence.
“In an era with widespread pneumococcal vaccination and low prevalence of bacteremia in the United States,noted Dr. Fritz and associates.
Dr. Fritz had no conflicts of interest to report. Some coauthors cited multiple sources of potential conflict of interest related to consulting fees, grant support, and research support from various pharmaceutical companies and agencies. The study was funded by the National Institutes of Health and in part by a grant from the National Institute of Allergy and Infectious Diseases.
SOURCE: Fritz C et al. Pediatrics. 2019;144(1):e20183090.