Background: Fresh-frozen plasma (FFP) is widely used in critically ill patients, despite a weak evidence base. Factors that influence the decision to transfuse FFP before intravascular catheter insertion are poorly described.
Methods: We undertook a case-controlled study based on a prospective cohort study of 1923 admissions to 29 intensive care units in the UK. Non-bleeding patients with an international normalized ratio (INR) ≥1.5 who underwent intravascular catheterization, but no other invasive procedure, were identified. We compared patient characteristics, illness-related factors, and biochemical and haematological variables between patients who did or did not receive pre-procedural FFP.
Results: One hundred and eighty-six patients fulfilled the criteria; 26 received FFP during the 24 h before line insertion (cases) and 160 did not (controls). Factors associated with greater use of prophylactic FFP by clinicians were pre-existing chronic liver disease (P=0.01), higher serum bilirubin before procedure (P=0.01), lower platelet count (P=0.01), higher activated partial thromboplastin time (P=0.001), lower fibrinogen (P=0.01), and concurrent red cell transfusion despite the absence of bleeding (P=0.001). There was no difference in pre-procedural INR [median (1st, 3rd quartile) cases: 1.95 (1.85, 2.6); controls 1.8 (1.6, 2.3); P=0.19]. The mean FFP dose was 11.1 ml kg(-1) (sd 5.7 ml kg(-1)); 53.8% of cases were transfused <10 ml kg(-1).
Conclusions: Chronic liver disease and more abnormal coagulation tests were associated with greater probability of pre-procedural FFP administration before vascular catheterization, whereas the severity of prothrombin time prolongation alone was not. FFP was more likely to be administered when red cells were also transfused, even in the absence of bleeding.