Chlorhexidine beats povidone-iodine in preventing vascular catheter colonisation

An impressive randomised multi-centre study in Japan has evaluated the efficacy of 0.5 and 1% alcohol/chlorhexidine with 10% povidone-iodine in preventing colonisation of vascular catheters. The study concluded that chlorhexidine at either concentration is superior to povidone-iodine in preventing the colonisation of vascular catheters.

The study was performed in 16 Japanese intensive care units. The 796 central venous or arterial catheters included in the final analysis were randomised to 0.5% and 1% alcohol/chlorhexidine with 10% povidone-iodine during insertion and dressing changes. Catheter-tip colonisation was 3.7 events per 1000 catheter-days in the 0.5% chlorhexidine group, 3.8 in the 1% chlorhexidine group, and 10.5 in the povidone-iodine group. The rate of catheter-tip colonisation was statistically significantly lower in both chlorhexidine groups compared with the povidone-iodine group.

It's important to note a couple of limitations in this study. Firstly, the authors compared chlorhexidine in alcohol with a povidone-iodine solution. So it isn't possible to disentangle the effect of alcohol from chlorhexidine using this design. Secondly, although the authors tested two concentrations of chlorhexidine (0.5% and 1%), both are lower than would usually be used in clinical settings in most parts of the world (2%). Thirdly, there was no impact on the rate of catheter-related bloodstream infections between the two groups, although this may be a factor of sample size, because the primary outcome, for which the study was powered, was catheter tip colonisation.

Despite these limitations, the findings of this large randomised trial are compelling, finding that vascular line-tip colonisation is less than half as likely when chlorhexidine is used during insertion and ongoing line care compared with povidone-iodine. This reinforces that chlorhexidine is a better choice than povidone-iodine for skin decontamination during line insertion and ongoing line care in the ICU setting.

jon_CHGvsPI_ICU_japan_25-1-18

JOIN THE DISCUSSION

 
Time to disinfect the loo before flushing?

toiletA fascinating new study in AJIC suggests that adding a disinfectant to the toilet bowl before flushing results in a significant reduction in viral contamination of a bathroom. Does this mean we should disinfect the loo before flushing it?

The study used MS2 coliphage as a proxy marker of pathogenic virus contamination. The degree of surface contamination of the bathroom was assessed with and without adding the phage to the toilet bowl (at a high concentration of 10^12) before flushing. The bathroom was heavily contaminated with the phage when it was added to the toilet bowl; surfaces were contaminated up to a concentration of 10^6 per 100cm2. Then, the experiments were repeated but this time disinfectants were added to the bowl before flushing: 5-10% hypochlorite, 0.5-2% hydrogen peroxide, QAC, or 0.23% peracetic acid. Perhaps surprisingly, the hypochlorite had limited impact on the concentration of phage in the toilet bowl until a contact time of 30 minutes, whereas both QAC and peracetic acid resulted in a ~2-log reduction with only a 1 minute contact time. The hydrogen peroxide made a limited impact on the concentration of phage at any concentration or contact time.

So does this mean we should disinfect the toilet bowl before flushing? The major problem with this is that the toilet bowl will be full of bodily fluids and organic matter (well, it's what the toilet is for), so there's a big risk that any attempt to disinfect the toilet bowl immediately before flushing are unlikely to reduce contamination of the bathroom. The author did attempt to simulate waste in the bowl using a microbiological culture broth - but I suspect the real thing would be a tougher challenge! However, this study does illustrate the risk of contaminating the bathroom with viral aerosol through toilet flushing, and argues for regular (perhaps more regular) disinfection of the toilet bowl!

JOIN THE DISCUSSION

 

Blog Archive