Rioux-Massé et al from University of Minnesota Medical Center report in the December 2014 issue of Transfusion (abstract) a retrospective study of 32 patients who had laboratory confirmation of immune-mediated PLT refractoriness, and received platelet transfusion with crossmatched or HLA-matched PLTs.
Immune-mediated refractoriness was called "positive" when a PLT crossmatch was incompatible with at least 1 unit or an HLA antibody screen detected HLA Class I antibody against any number of antigens in a single-well test system. The HLA antibody avoidance technique was not used by the blood supplier, ARC Mid-America Blood Services Division. Finally, PLTs were either ABO-identical or -compatible.
Out of 354 total PLT transfusions, 45% were random-donor apheresis PLTs, 42% were crossmatched PLTs, and 13% were HLA-matched. Using a CCI of 5.0 X 109/L as defining a satisfactory response to PLT transfusion, the rates of satisfactory responses for random-donor units was 13%, for crossmatched PLTs 25%, and HLA-matched PLTs 29%. There was no statistically significant differences in the frequency of satisfactory transfusion between the different PLT products, although this study was probably not adequately powered to detect any true differences. One of the interesting, albeit frustrating, things about this study is how miserably crossmatched PLT and HLA-matched platelets performed overall.
The authors did not correlate the panel-reactive antigen (PRA) with the crossmatch assay, since they only used a screening test for HLA antibodies. This is somewhat odd to me, since we use at Rush University Medical Center a Luminex screening assay for PRA and reflex to a more comprehensive test if the PRA is over 40%; I and the heme/onc clinicians are looking for the PRA to determine whether or not to pursue HLA-matching.
Another interesting thing is that patients in this study were tested for platelet-specific antibodies but this was not used to define immune-mediated PLT refractoriness nor administer HPA-negative PLT units, and, further, did not correlate with refractoriness. (Why were the patients tested for HPA antibodies anyway, then?)
The limitations of this study are that it is a small study and was not a randomized trial between random, crossmatched, and HLA-matched PLTs. However, this study adds to literature suggesting that clinicians should not expect any dramatic PLT increments from either crossmatched or HLA-matched PLT transfusions once refractoriness has been established. We really need a randomized controlled trial to definitely determine the best approach to PLT transfusion in this difficult to manage group of patients. Currently, much time and many resources are consumed chasing after HLA-matched PLTs, which generally do not appear to provide any more clinical benefit than random apheresis PLTs.