I am researching the optimal speed/duration when preparing autologous PRP. The PRP is for immediate reinjection in the clinical setting. I am finding that the supernatant platelet concentration is higher than expected and is reduced by >80% in the residual plasma in the RBC fraction. I cannot find any reference to these unexpected results in the literature.
In summary I have been spinning blood in 10ml citrated tubes at various speeds from 100g to 200g for 5 and 10 mins and measuring the platelet levels in the supernatant plasma with a Coulter Counter. When analysing the results I was finding that the increased concentration of platelets in the supernatant was more than theoretically possible (over and above that due to removal of the RBCs). If this was not an artefact then it suggested that the platelets concentration in the residual plasma still within the RBC layer after centrifugation is depleted, and the because some or all platelets are displaced into the supernatant layer, causing an unexpectedly higher increase in supernatant platelet concentration
I have done several more refined runs where I:-
- measured the number and concentration of platelets in the remaining plasma within the centrifuged RBC layer and found them reduced by up to 80-90% compared to the supernatant plasma
- compared the total number of platelets in uncentrifuged citrated blood against the total of platelets in all layers of the centrifuged citrated blood and found them to be the same
I am wondering whether this effect is merely related to fluid dynamics of of small and large particles in suspension. In the previous question there was the suggestion that it might be related to plasma viscosity. My feeling is that this is unlikely in explaining why platelets are being transferred upwards against the effects of g-force, and is anyway present in multiple samples from different donors
The ramifications of this are important since it shows that higher concentrations of PRP within the supernatant can be gained by reducing the speed of centrifugation to below 150g. The level of unexpected increase rises as the speed drops, partly because the proportion of plasma remaining in the RBC fraction rises (with associated displacement of the platelets into the supernatant) and partly because fewer platelets are lost to the buffy layer.
I have not found this published anywhere but wonder how many other researchers have seen this phenomenon. Have I missed mention of it in previously published work? Two researchers commented in the previous question that they had seen it as an incidental finding when using small volumes, but this is still present in 10 ml tubes of human blood where there are no practical limitations on the quantities of blood needed for this research.
What kind of a platelet count do you get? When we prepared PRP of healthy volunteers by 4 min centrifugation at 500 rcf the average (+- standard deviation) platelet count was 420 +- 90 x10^9/L.
The final concentration of PRP will depend on both the initial platelet count and the haematocrit. with a haematocrit of 0.5 then one would expect a theoretical increase in platelet concentration of 2x merely by removing the RBCs, and this was what I was expecting to measure. However I was getting increases in concentration of up to 3x in the supernatant from a single spin and this lead to my further research
What size tubes were you using?
We have noticed this before with platelet production from buffy coats, but the more platelets you recover, the more contamination with red cells and leukocytes you will have. For the purpose of treatment with autologous PRP this might be no problem (also not supported by data), but for earlier production of PRP for diagnostic tests I guess that a compromise was taken between getting enough platelets for the test, with a reasonable cellular contamination/volume.
PRP plasma is often gradient dependent and gravity independent and this could provide your answer.
Please see attachments.
Working on platelet aggregation of patient with macrothrombocytopenias I usually decrease the g force when I expect large platelet ( 80-120g). Try to see if different donors have different platelet volume .
Macrothrombocytes are usually young platelets as are often seen in disorders such as idiopathic thrombocytopaenic purpura. In Essntial thrombocytosis one would expect a thrombitic tendency but the platelets are usually defective and fail to aggregate to normal aggregating agents !
We prepared PRP from mice blood and we observed similar trends. The platelet count increases near the red layers when we use with low g forces to separate platelets.
I suspect this could be due to the different sizes of platelets. I think younger platelets are large with high MPV. With lower centrifugal force, RBCs alone get sedimented, while at higher force, RBC and large platelets are down. That could be one explanation. To clarify this, MPV can be monitored.
In case, you think viscosity brings the platelets down, I have experience diluting whole blood 1:1 with PBS and see enriched platelets in PRP.
Its important to have round bottom tubes and centrifuge without brakes to reduce aggregates of platelets, which will reduce the platelet number in the PRP.
Hi Hariharan
I agree with you that viscosity by itself is unlikely to be important, though the relative densities will be. However the issue here is that there seems to be an active process that is displacing the platelets from the remaining plasma within the RBC layer sinxe platelets are more dense than plasma.
Assuming the experiential observations are true what forces might they be?
I must say it's strange!. We have not done it systematically and also the work was not for clinical work but rather looking at the effects of platelets on formation of autoagglutinates in malaria infection. what i remember is we used to separate plasma from RBC at 1800 rpm for 10min. Then we further spin the plasma to make PPP and PRP at 3000rpm for 15 min and it worked well for us we could also confim with giemsa stained slides and coulter counter.
What i did not confirm is when we received a donor who was lipaemic. I would think there may be differences but am not sure.
Hi Chris
Did you ever find out more about your observations? I am looking to get the best yielding PRP (for hair loss) Can you possibly assist me with a two spin method? Manytahnx
Hi Chris
Did you ever find out more about your observations? I am looking to get the best yielding PRP (for hair loss) Can you possibly assist me with a two spin method? Manytahnx
Hi Larry
As a rule of thumb, if you fill a 10ml blood tube of blood containing 0.5ml 3.8% sodium citrate as an anticoagulant at 150g for 10 mins then you will get a reasonable amount of platelet rich plasma containing no RBCs and very few neutrophils. The volume of plasma will depend on the haematocrit.
There is no consensus on whether you need to reverse the citrate with calcium chloride as the plasma will clot with all the collagen present
Thanx so much Chris
Once I have done the 10 min spin (soft at 150g) should I remove all the plasma and do a second spin at say 400g (hard)for 15 min to further concentrate the PRP??
- Badges
- Science topic
- Similar topics
- Biological Science
- Anatomy
- Tissue