Value of superficial cultures for prediction of catheter-related bloodstream infection in long-term catheters: a prospective study

Abstract

Cultures taken from the skin and from the hubs of short-term central venous catheters can help us to predict catheter-related bloodstream infections (C-RBSIs). The value of these cultures for such predictions has not been assessed in long-term catheters. Our objective was to assess the value of superficial cultures for the prediction of C-RBSI among patients with long-term catheters. Over a 2-year period, we prospectively obtained cultures from the skin overlying reservoir ports (group A) and from the skin insertion site and hubs of all tunneled catheters (group B). This routine was performed by vascular and interventional radiologists immediately before catheter removal (irrespective of the reason for withdrawal). Swabs were processed semiquantitatively. Catheter tips from both groups were cultured using Maki's semiquantitative technique and sonication. We also performed cultures of the reservoir ports at different sites. C-RBSI was defined as the isolation of the same species of microorganism(s) both in the colonized catheter and in at least 1 peripheral blood culture. We included 372 catheters (group A, 223; group B, 149) during the study period. The catheter colonization rate was 23.4% (87/372), and 28 patients had C-RBSI. Validity index values for the capacity of surface cultures to predict C-RBSI in groups A and B were, respectively, as follows: sensitivity, 23.5% and 45.5%; specificity, 59.7% and 63.0%; positive predictive value, 4.6% and 8.9%; and negative predictive value, 90.4% and 93.5%. Superficial cultures of patients with long-term catheters could help us to rule out the catheter as the portal of entry of bloodstream infections. Superficial cultures (from skin and hubs) proved to be a useful conservative diagnostic tool for ruling out C-RBSI among patients with long-term tunneled catheters and totally implantable venous access ports.