A study published last year looked at this very question!
It was a well powered (90 patients randomized) level 1 study comparing results using PRP versus normal saline.
3 injections at monthly intervals. Averaged a 38cc blood draw, two spin system
3cc LP-PRP injected (no further quantification of PRP).
6 month Results:
VAS Pain Score (10 point scale, lower is less pain, results were statistically significant)
PRP group decreased from 8.0 to 5.7
Normal Saline group from 7.9 to 6.6
WOMAC (lower score is better function, results were statistically significant)
PRP group from 82 to 60
Saline group 78 to 71.
From evaluating the authors' PRP blood draw volume and centrifuge technique....the PRP dose was probably in the range of 3-6 Billion per injection.
I draw at least 60cc for every knee and inject 8 cc so the dose per injection is much higher....on average 10 billion platelets..and the higher dose results have been shown in at least one study to be better. (H. Bansal et al. in 2021)
You can measure a protocol's clinical efficacy not only by patents' decrease in pain and WOMAC but also by needing fewer injections or getting clinical benefit for longer than 6 months.....all of which I see with higher dose LP-PRP for knee osteoarthritis.
This is a very good high level study from some thoughtful Indian researchers.
We have an increasing volume of clinical evidence that PRP injections can offer some hope of pain relief even to patients with advanced knee arthritis.
It is up to the patient to decide on this treatment modality after discussing relevant surgical and nonsurgical treatment options with their orthopedic trained doctor.