Perioperative blood conservation strategies: weighing the medical evidence — I

To the Editor: We read, with interest, the recent editorial written by Drs. Karkouti and McCluskey.1 The crux of their editorial is that cardiac anesthesiologists should discontinue most blood conservation methods to avoid transfusions in heart surgery patients, until such methods have been proven safe and efficacious. However, the editorial leaves a number of unanswered questions; most importantly, is receiving blood, or blood component therapy, harmful? The authors suggest a causal relationship between the volume of transfusion and mortality. Work in this area has been based upon retrospective, propensity scored, database analysis which describes an association between transfusion and mortality from small to large volume transfusion.2–4 Possibly the nature of the surgical insult, with a requirement for large volume transfusion, is the major determinant for excess mortality, not the consequent transfusion. The findings of well-controlled trials, which examine simple methods to reduce transfusion requirements, can potentially be dismissed, arguing that they are not powered adequately to detect differences in mortality. However, it should be recognized that these studies were frequently not intended to evaluate mortality. There is enormous variation in transfusion practice across centres performing the same type of cardiac operations.5 Many studies have evaluated methods that might reduce this variation in practice. Simple organizational changes and audit of practice may be just as effective in reducing transfusion rates as using mechanical or pharmacological methods. We have a safe blood supply; but this editorial largely ignores the wider issues around an increasingly scarce and expensive product. In addition, preventing low volume transfusions may be important to functional outcome after cardiac surgery.6 In the void left after the closure of the Blood conservation using antifibrinolytics: a randomized trial in a cardiac surgery population study, have Karkouti and McCluskey stopped using tranexamic acid in patients not involved in randomized clinical trials? Are they suggesting that units should stop using standard tests of coagulation in patients, as the value of these tests have not been validated in hemodiluted post-cardiopulmonary bypass coagulopathic patients? The authors have produced a stimulating editorial. To do no harm, we must first correct the balance between transfusion and conservation approaches. This will require primary and database research into transfusion of blood, use of component therapy, and mechanical and pharmacological methods to reduce transfusion. At a time when aprotinin has been removed from our armamentarium, abandonment of a range of methods of transfusion avoidance may not be timely.