As in many aspects of transfusion medicine, practices have developed based on low levels of evidence. Let’s review some assumptions and think how these might change in future:- (1) Indications. 80% of platelet transfusions are given prophylactically to patients with bone marrow failure, yet no trials have demonstrated the value of prophylaxis over treatment of bleeding alone. (2) Trigger. A trigger of 10 × 10 g/l is now accepted. Reducing this further will depend on accurate counting, challenging at this level. (3) Dose & frequency. The optimum dosing schedule has not been derived by pharmacokinetic studies, but by what can be conveniently manufactured. An adult dose of 3 × 1011 platelets is accepted, but there is wide variation in practice. (4) Manufacture. Globally, 3 methods (pooled buffy coats, platelet rich plasma and apheresis) all produce an efficacious product. Although apheresis platelets reduce donor exposure, no functional advantage has been demonstrated. (5) In vitro assessment of function. A battery of tests is recommended, but only pH has been shown to correlate with posttransfusion survival. (6) ABO matching. ABO compatible platelets give better increments, and transfusing ABO incompatible plasma can be associated with haemolysis, but ABO matching to patient is often compromised by stock management. (7) Use of platelet additive solution. A 70:30 mix of PAS:plasma would reduce reactions and perhaps TRALI but this is not standard practice. Is the plasma useful? (8) CMV safety. Overall, leucocyte depletion alone appears to provide a high degree of protection, but this is not universally accepted. (9) Assessment of response. 1 and 24 hr increments are standard, but is regression analysis the way forward? (10) Management of refractoriness. Better HLA and HPA matching is now technically possible. Should these lead to better HLA matching for all? (11) Is one product for all appropriate? 20% of platelets go to surgical patients who are bleeding. Do they need a product with a different specification?
Read full abstract