Drs Wolfe and Kress coauthored a terrifically clinically useful and scholarly assemblage of common ICU and related procedures and their hemorrhagic risks, as well as thoughtful approaches to reducing risk in a recent issue of CHEST (July 2016). However, I think their analysis regarding, and view that, preprocedural blood product or other infusions to reduce bleeding risks are often wrong, is wrong itself. Here’s why:
They cite (their reference 56) the sole prospective controlled study of preprocedure transfusion as showing “no difference in bleeding complications in those who underwent transfusion prophylactically and those who did not.” Here’s what the article’s authors concluded instead: “Due to limited inclusion, noninferiority of omitting FFP transfusion could not be demonstrated.” There was only one major bleed among the 81 patients enrolled. The study was stopped early due to poor accrual. Patients were excluded from that study if they’d received warfarin, which is often exactly the type of patient ICU physicians want to treat with a hemostasis-improving infusion before a procedure.
Drs Wolfe and Kress note that after infusions of fresh frozen plasma, for example, blood tests like the prothrombin time often fail to improve significantly. But the point is not to fix the blood test, it is to reduce patient risk. Replenishing depleted factors temporarily can reduce bleeding risk and improve the safety margin. As Dr Baron explained in a recent thoughtful and well-documented Point/Counterpoint article in CHEST that focused on coagulopathy and central venous line insertion, “Why take chances?”
Volume overload (particularly in anuric patients) and further lung injury from fresh frozen plasma infusion is certainly a concern. But lack of proof of benefit, based on meta-analyses of case series, is not. Ascertainment of bleeding in these series is not assiduous, and as a leading pulmonary physician pointed out decades ago, “Absence of evidence is not evidence of absence.”