In our experience, the incidence and pathogenesis of colonization may
vary in different types of catheters. Arterial Swan-Ganz, and hemodial
ysis catheters are good examples of this problem. This observation has
implications for diagnosis, and the best method for each type of cath
eter may not be the same. In our opinion, laboratory diagnosis in dail
y practice should be limited to external surface cultures of the tip o
f intravascular catheters. Intraluminal cultures should be limited to
research purposes, except in catheters used for parenteral nutrition o
r hemodialysis. In this case, the intradermal segment gives more sensi
tive information. Sensitivity in diagnosing pulmonary artery colonizat
ion can be improved by evaluating both the tip and intradermal segment
s. In presence of an indwelling introducer the intradermal segment sho
uld be replaced by the introducer tip. (C) Elsevier Science Inc. 1997.