Diagnosis of catheter-related bacteraemia: a prospective comparison of thetime to positivity of hub-blood versus peripheral-blood cultures

Background A method of diagnosing catheter-related infection (CRI) without removing the catheter would be useful. An earlier positivity of central com pared with peripheral venous-blood cultures may be associated with catheter -related bacteraemia. We evaluated prospectively the differential time to p ositivity (DTP) of paired blood cultures drawn simultaneously via the cathe ter hub and from a peripheral venous site. Methods Over a 14-month period in an intensive-care unit of a cancer referr al centre, simultaneous hub-blood and peripheral-blood cultures (a mean of two per patient) were obtained from patients with a suspected CRI. Accordin g to clinical criteria and quantitative culture of the catheter tip, cases were classified as CRI or sepsis of other origin. At least one pair of hub- blood and peripheral-blood cultures was obtained within 48 h before cathete r removal, and we recorded the DTP between hub-blood and peripheral-blood c ultures with an automatic device for detection of blood culture positivity. Findings We analysed 93 catheters removed because of suspicion of CRI, in 2 8 episodes, the same micro-organisms were found in both hub-blood and perip heral-blood cultures. A diagnosis of definite bacteraemic CRI was made in 1 6 of the 17 patients in whom a positive hub-blood culture was detected at l east 2 hours earlier than peripheral-blood culture. About half (9/17) of th ese episodes occurred in long-term (>30 days) devices. CRI was excluded in ten of the 11 patients with a DTP lower than 2 h. The DTP of paired blood c ultures was significantly greater in patients with CRI than in others (p<10 (-4)). A cut-off DTP value of 120 min had 91% specificity and 94% sensitivi ty for the diagnosis of CRI. Three of 17 episodes with only hub-blood cultu re positive were associated with CRI. Interpretation This prospective study suggests that measurement of the diff erential time to positivity between hub-blood and peripheral-blood cultures is a simple and reliable tool for in-situ diagnosis of catheter-related se psis in cancer patients. Further studies are needed to confirm these data f or short-term catheters.