Central venous catheters—time for a change?

It is often difficult to ascertain exactly where a particular medical practice or policy originates. Thus it is with routine scheduled changes of central venous catheters in patients requiring prolonged intensive care. If we are to believe the telephone survey by Cyna et al in this week’s issue (p 1944),¹ the policy of routinely replacing central venous catheters to reduce a perceived high incidence of catheter related sepsis appears to be ingrained in many British intensive care units. That this policy continues is surprising, since it is impossible to find a published randomised trial in the past 12 years supporting the contention that the incidence of catheter related sepsis increases with duration of catheterisation.

Indeed, there are now several randomised controlled trials comparing routine catheter change with change when clinically indicated, and these have been the subject of a recent meta-analysis.² Although routine exchange over a guidewire is associated with fewer technical complications but a higher incidence of catheter colonisation and infection than routine replacement at a new site, neither approach confers any benefit when compared with changes only when clinically indicated. How recently had those units stating that their routine change policy was based on published evidence actually reviewed the literature, or are they confused with the evidence on pulmonary artery catheters—which should be removed within 72 hours of insertion?³ A recent study has shown that the incidence of systemic infection associated with pulmonary artery catheterisation can be reduced to zero if the catheter is removed within 4 days.⁴

Routine change of central venous catheters is expensive and has appreciable morbidity and, in critically ill patients, potential mortality. Clearly, intensive care physicians supporting such a policy must perceive that there is a risk-benefit advantage. Is such a view justified? Published studies suggest that the risk of catheter colonisation is around 25% and of infection 5%^5–6—levels far lower than most of us might have expected. Why is there this inaccurate perception of the size of the problem? Is it because outside trials, in routine clinical practice, the incidence of infection is far higher? Maybe it is because this nosocomial infection is by definition iatrogenic and leads to feelings of responsibility and guilt. The latter seems unlikely since many doctors have failed to accept the convincing evidence that use of maximal aseptic technique (sterile surgical field, surgical mask, gown, and gloves) at insertion significantly reduces infective complications.⁷

As multilumen catheters are no more likely to become infected than single lumen devices and have the additional benefit of reducing the need for peripheral venous access (just as susceptible to infection),⁸ the days of the “no touch, no glove” insertion of single lumen internal jugular lines so beloved of cardiothoracic anaesthetists should be long past. Finally, it should be remembered that, although a patient with catheter related sepsis may present seriously ill with severe haemodynamic disturbance, removal of the catheter is often the only treatment needed, and these infections generally run a relatively benign course.⁹

It appears that many British intensive care units need to review their policies on inserting and changing central venous catheters. As well as improving insertion technique and abandoning the routine change, doctors should also consider other changes supported by evidence from randomised clinical trials: the use of povidone-iodine ointment and cotton gauze dressings at insertion⁶ and of catheters impregnated with antibiotics or antiseptics. One study has shown a reduction in the incidence of colonisation to 13.5% and of infection to 1% with a catheter impregnated with chlorhexidine and silver sulphadiazine.⁵ Once the catheter is inserted, the environmental factors shown to reduce colonisation and infection are provision of adequate nursing and medical staffing levels,¹⁰ use of special teams for catheter care, and avoiding excessive manipulation of the catheter.⁶ It might be interesting if Cyna and colleagues were to repeat their study in, say, a year’s time so that we can see if the message has finally filtered through. More likely, the intensive care community will be so fed up of answering these time consuming telephone calls that we shall never know.

Papers p 1944