Background: Paired quantitative and qualitative blood cultures have been introduced for the diagnosis of catheter-related bloodstream infections (CRBI) with the catheter in situ. The aim of the study was to compare the diagnostic performance and the prognostic value of the two methods in the evaluation of febrile episodes without an apparent source in children with cancer.
Procedure: During a 4-year period, in every febrile episode without an apparent focus, blood was drawn simultaneously from the catheter lumen and a peripheral vein in order to perform paired quantitative (Isolator) as well as qualitative (BacT/Alert) blood cultures. The diagnosis of a CRBI was defined as either a case of greater (at least 10 fold) or earlier (differential time to positivity >2 h) bacterial growth from the catheter compared to the peripheral blood sample, respectively.
Results: Nineteen febrile episodes manifested in 16 children (total period of observation 11,150 catheter-days) were evaluated with both methods. A concordant diagnosis of CRBI was stated with both methods in six episodes; one episode was diagnosed as CRBI only with qualitative culture criteria. Treatment failure resulted in catheter removal in five out of the seven episodes defined as CRBI with either method. Episodes where a CRBI was ruled out with both methods had a favorable outcome.
Conclusions: In this study the two methods showed comparable results in the diagnosis of CRBI and both were of prognostic significance, regarding the outcome of the treatment. However, large scale studies are required in order to evaluate the clinical relevance and the cost effectiveness of performing routinely paired blood cultures with either method.
2005 Wiley-Liss, Inc.