Consistency between guidelines and reported practice for reducing the risk of catheter-related infection in British paediatric intensive care units.

PURPOSE: Optimal strategies for reducing catheter-related blood stream infection (CR-BSI) differ for adults and children. National guidelines do not make child-specific recommendations. We determined whether evidence explained the inconsistencies between guidelines and reported practice in paediatric intensive care units (PICUs). METHODS: We conducted a survey of eight interventions for reducing CR-BSI in all 25 British PICUs in 2009. Interventions were categorised as requiring child-specific evidence, generalisable to adults and children, or organisational recommendations. RESULTS: Twenty-four of the 25 PICUs responded. For child-specific interventions, practice diverged from guidelines for "Insert into subclavian/jugular veins" (18 PICUs frequently used femoral veins, supported by observational evidence for increased safety in children). Practice reflected guidelines for "Use standard but consider antimicrobial-impregnated central venous catheters (CVCs) for high-risk patients" (14 used standard only, 3 used standard and antimicrobial-impregnated despite no randomised controlled trial (RCT) evidence for antimicrobial-impregnated CVCs in children, 7 used heparin-bonded for some or all children); "Use 2% chlorhexidine for skin preparation" (20 PICUs); "Avoid routine CVC replacement" (20 PICUs). For generalisable interventions, practice was consistent with guidelines for "Administration set replacement" (21 PICUs) but deviated for "Maintenance of CVC asepsis" (11 PICUs used alcohol due to inconclusive evidence for chlorhexidine). Practice diverged from guidelines for organisational interventions: "Train healthcare workers in CVC care" (9 PICUs); "Monitor blood stream infection (BSI) rates" (8 PICUs). CONCLUSIONS: Guidelines should explicitly address paediatric practice and report the quality of evidence and strength of recommendations. Organisations should ensure doctors are trained in CVC insertion and invest in BSI monitoring, especially in PICUs. The type of CVC and insertion site are important gaps in evidence for children.