This article is more than 5 years old. Information may no longer be current.
Model predicts bloodstream infection risk in febrile pediatric cancer patients
Click Here to Manage Email Alerts
A diagnostic model accurately predicted the probability of bloodstream infections among clinically stable, febrile pediatric oncology patients, according to a recent study.
“Guidelines for hospital admission and use of empiric intravenous antibiotics exist for febrile patients who are severely neutropenic or ill appearing. [However,] management of pediatric oncology patients without severe neutropenia is less clear,” Adam J. Esbenshade, MD, MSCI, of the department of pediatrics at Vanderbilt University School of Medicine, and colleagues wrote. “If the diagnosis of bacterial blood stream infections (BSI) could be predicted with precision, an evidence-based algorithm could be established and incorporated into clinical care to guide judicious empiric antibiotic therapy.”
To build and validate a diagnostic prediction, researchers identified a cohort of 463 patients diagnosed with childhood cancer between 2007 and 2010. All patients were aged younger than 23 years at diagnosis.
The researchers identified episodes of fever among these patients between 2007 and 2012 while using a central venous catheter (CVC) and with an absolute neutrophil count (ANC) of at least 500 µL. Fever was classified as a temperature over 38°C for more than1 hour or 38.3°C for any period of time.
The researchers used a data collection instrument to extract patient demographics and treatment and symptom-based risk factors available within 2 hours of fever. Components used to build the prediction model were: ANC, absolute monocyte count, body temperature, inpatient/outpatient observation, gender, CVC type, hypotension, chills, cancer diagnosis, stem cell transplant, upper respiratory symptoms, and cytarabine, anti-thymocyte globulin or anti-GD2 antibody exposure.
Of the 932 fever episodes among patients, there were 91 cases of BSI. Independent predictors for infection included:
- higher body temperature (OR = 2.36; 95% CI, 1.693-3.293);
- tunneled external catheter (OR = 13.79; 95% CI, 6.587-28.862);
- peripherally inserted central catheter (OR = 3.95; 95% CI, 1.529-10.21);
- elevated ANC (OR = 1.19; 95% CI, 1.023-1.385);
- chills (OR = 2.09; 95% CI, 1.068-4.094); and
- hypotension (OR = 3.08; 95% CI, 1.421-6.7).
Reduced risk for BSI was associated with diagnosis of acute lymphoblastic leukemia (OR = 0.34; 95% CI, 0.131-0.879); increased age (OR = 0.7; 95% CI, 0.489-0.998) and drug exposure (OR = 0.08; 95% CI, 0.029-0.211).
“Our model, which demonstrates accurate probability estimation for bloodstream infection in patients with CVCs, can contribute to the development of an evidence-based algorithm, which can be incorporated into clinical practice, whereby parameters can be set to guide antibiotic management,” the researchers wrote. “In such a model, one may consider no antibiotics for low-risk, outpatient antibiotic administration for intermediate probability, with choices based on local antibiogram, and admission with broad-spectrum coverage for high-probability patients.” – by Jen Byrne
Disclosure: The researchers report no relevant financial disclosures.
Click Here to Manage Email Alerts