SAN DIEGO – Obtaining blood cultures in children hospitalized for community-acquired pneumonia led to longer hospital stays and caused physicians to order more broad-spectrum antibiotics, results from a retrospective cohort study showed.
The 2011 Pediatric Infectious Diseases Society/Infectious Diseases Society of America Community Acquired Pneumonia guidelines included a strong recommendation to obtain blood cultures for all children admitted with moderate to severe CAP, “but that was based on low-quality evidence,” Dr. Michael P. Koster said in an interview at the annual meeting of the American Academy of Pediatrics. “Very little is known about how blood cultures influence the management of pneumonia. Our question was getting at whether or not getting a blood culture changes practice management.”
Doug Brunk/Frontline Medical News
Dr. Michael Koster
To investigate, Dr. Koster and his associates at four separate medical institutions retrospectively evaluated the charts of 1,142 children aged 3 months to 18 years who were hospitalized for community-acquired pneumonia (CAP) during 2011 and 2012 according to ICD-9 codes for CAP or respiratory distress. Children with severe medical comorbidities were excluded. Dr. Koster, a pediatrician at Hasbro Children’s Hospital, Providence, R.I., and his associates collected data on patient demographics, medical history, laboratory tests, diagnostic radiography, antimicrobials administered, length of stay, ICU transfer, and readmission.
Of the 1,142 initially identified, 763 were used in the final analysis. Of these, 462 had blood cultures and the remaining 301 did not.
Dr. Koster reported that patients in the blood culture group had a significantly longer mean length of stay, compared with the no culture group (3.4 vs. 1.9 days, respectively; P < .0001). This difference persisted when ICU patients were removed from the analysis (2.5 vs. 1.8 days; P < .0001).
The researchers observed no statistically significant differences between those who had blood cultures and those who did not in receipt of antibiotics prior to presentation (41.6% vs. 40.9%, respectively; P = .85), the presence of any pulmonary effusion (57.6% vs. 61.1%; P = .33), or in the 14-day readmission rates (4.1% vs. 3%; P = .42).
However, in the emergency department, the blood culture group was more likely receive a third-generation cephalosporin (68.9% vs. 42.9%; P < .0001) while the no culture group was more likely to receive penicillin/ampicillin alone (38.4% vs. 21.3%; P = .0001).
Among patients in the blood culture group, 2.4% were positive for pathogens and 2.2% were contaminated, for a positive predictive value of 52.4%.
“Blood cultures aren’t free of harm,” Dr. Koster concluded. “They increase how long you stay in the hospital, and they increase the prescription of third-generation cephalosporins. Practice variation among physicians is probably what resulted in these findings. Until we recognize the variance around the interventions that we give to kids, we won’t be able to decrease the variance around their outcomes.”
The study was supported by the Rhode Island Foundation. Dr. Koster reported having no financial disclosures.
On Twitter @dougbrunk