Infection v. colonisation.

The interaction between bacteria and the human respiratory tract is complex and while the concept of three states, namely sterility, colonisation, and infection is clinically convenient it is inevitably in oversimplification. Evidence from both clinical and laboratory observations has led to some ideas about the relationship between colonisation and infection and while these are helpful in defining the steps involved, the decision of whether and when to start new treatment remains one of clinical judgement. This article reviews the evidence from lung disease both in and out of an intensive care unit and attempts to define the frontier between infection and colonisation in different clinical settings.