If there is more than minimal erythema or ANY purulence at the exit site, the catheter is likely infected. It should be removed and a new one replaced at a different site.
Lack of erythema does not exclude catheter infection
Two sets of blood cultures should be drawn with AT LEAST one (and preferably both) from peripheral sites. Blood cultures drawn through non-tunneled catheters are more likely to yield contaminants. One set of cultures may be drawn through a catheter if it is tunneled.
The utility of cultures of the catheter tip itself is not well defined, and should NOT routinely be obtained when lines are removed.
All patients should be followed closely, and repeat cultures should be sent if clinically indicated.
When a catheter-associated BSI is associated with catheter dysfunction, consider the possibility of suppurative thrombophlebitis.