Objective: Central venous catheter (CVC)-related infections may be caused by micro-organisms introduced from the skin surface into deeper tissue at the time of CVC insertion. The optimal disinfection regimen to avoid catheter-related infections has not yet been defined. This study compares three different approaches.
Design: Prospective randomised trial.
Setting: A tertiary care hospital.
Patients and participants: One hundred nineteen patients scheduled electively to receive 140 CVCs.
Interventions: Skin disinfection was performed with either povidone-iodine 10% (PVP-iodine), chlorhexidine 0.5%/propanol 70%, or chlorhexidine 0.5%/propanol 70% followed by PVP-iodine 10%. Prior to disinfection, a swab from the site of insertion was taken for culture. CVCs were removed if no longer needed or infection was suspected. All catheters were cultured quantitatively after removal.
Measurement and results: Bacteria could be isolated from 20.7% of the catheter tips. Bacterial growth was found in 30.8% of the catheters placed after skin disinfection with povidone-iodine, in 24.4% after disinfection with propanol/chlorhexidine and in 4.7% after disinfection with propanol/chlorhexidine followed by povidone-iodine ( p=0.006). In 15 cases, the same organism was isolated from the skin swab and the catheter tip. Ten of these paired isolates showed the same pattern in a pulsed-field gel electrophoresis analysis.
Conclusions: Skin disinfection with propanol/chlorhexidine followed by PVP-iodine was superior in the prevention of microbial CVC colonisation compared to either of the regimens alone. These results support the concept that catheter infections can originate from bacterial translocation at the time of catheter insertion.