Hi Lauren and Julia,
I would just like to give this thread some nuances.
We have tested running whole blood using our standard protocol (as if we would run PBMCs) and it does not work well. Besides the problems Julia mentioned (the red blood blood cells are just so numerous, they block the way for the analyte to reach the capture antibodies), there will also be a lot of background coloring due to the blood which makes it hard to analyse possible spots.
I’m not aware of any systematic testing that’s been done to show at what concentration of red blood cells they start to interfere. But theoretically, one could imagine using separated pure PBMCs and then adding increasing amounts of Ficoll pellet (containing red blood cells and granulocytes) and study at what point problems arise. This would give some information on how many contaminating erythrocytes one can accept. Once there, one could take the sample where problems began and measure the haemoglobin value. That value could then be related to a “real” sample, to be able to say “ok, the haemoglobin value in this sample is under X, then it’s ok to run the assay”. But as said, we have not tried this in-house. Instead, to be on the safe side, we recommend doing a Ficoll separation, because then you know there will be minimal amounts of red blood cells there. An alternative or complement to Ficoll is a red blood cell lysis protocol. With some samples there is a risk of getting remaining contaminating red blood even after a Ficoll. I used to work with tonsils back in the day, and the PBMC band was always super red after Ficoll separation. I then complemented with a few minutes of red blood cell lysis solution which gave me a totally pure PBMC preparation. But if you’re working with fresh blood, there should be no need for other methods than Ficoll.