Background: Blood components are often given prophylactically before t
he placement of invasive lines in patients with coagulation defects. L
ittle, however, is known about the epidemiology of defects in these pa
tients. The purpose of this study is to ascertain what proportion of i
ntensive care patients who receive invasive lines have hemostatic defe
cts, what actions are taken by physicians to correct these abnormaliti
es before invasive line insertion, and what the incidence is of bleedi
ng complications after invasive line placement. Study Design and Metho
ds: Charts were retrospectively reviewed for 490 intensive care patien
ts in whom 938 arterial, pulmonary artery, and central venous lines we
re placed. Results: At least one defect in hemostasis was documented f
or 388 patients (41%) before line placement, with 253 (27%) of these p
atients evidencing severe abnormalities. Seventeen percent of patients
had no preprocedure laboratory evaluation, Trauma patients showed the
highest numbers of abnormalities in hemostatic testing, but medical p
atients had more-severe defects. The occurrence of isolated abnormal l
aboratory values did not predict bleeding, but a score derived from a
consideration of multiple defects did. Correction of the abnormalities
was attempted in 37 percent of patients with hemostatic defects. Sixt
een patients had bleeding complications, but only two had complication
s that were life-threatening. None of the complications were fatal. Co
nclusion: Invasive lines are used frequently in patients with hemostat
ic defects, often without any attempt to correct the abnormalities. Ne
vertheless, rates of hemorrhage are low and appear to be closely relat
ed to the level of experience of the physician rather than to defects
in hemostasis. These findings suggest that the use of blood components
for preprocedure correction of hemostatic defects is not necessary, e
xcept in those patients who have the most severe hemostatic abnormalit
ies.