Centrifugation Determines What Is in the Syringe, Not What Happens to the Patient: A Systematic Meta-Analysis of Platelet-Rich Plasma Preparation, Platelet Integrity, Cell Counts, Growth Factors and Clinical Outcomes in Regenerative Medicine
Márcio Hiroaki Kume¹*, Bianca Furlan², Camila Gobatto Boaventura², Mônica Andréa Probst², Edson Peracchi² and Carmen Austrália Paredes Marcondes Ribas3
1Sugisawa Hospital, Department of Regenerative Medicine, Curitiba, Brazil
2CeUnina, Department of Biologic Science, Curitiba, Brazil
3Mackenzie University, Curitiba, Brazil
*Corresponding author: Márcio Hiroaki Kume, 80250-190, Iguassu Avenue, 1236, Sugisawa Hospital, Department of Regenerative Medicine, Curitiba, Brazil
Citation: Kume MH, Furlan B, Boaventura CG, Probst MA, Peracchi E, et al. Centrifugation Determines What Is in the Syringe, Not What Happens to the Patient: A Systematic Meta-Analysis of Platelet-Rich Plasma Preparation, Platelet Integrity, Cell Counts, Growth Factors and Clinical Outcomes in Regenerative Medicine. J Stem Cell Res. 7(3):1-31.
Received: August 09,2026 | Published: August 21, 2026
Copyright©2026 by Kume MH, et al. All rights reserved. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
DOI: https://doi.org/10.52793/JSCR.2026.7(3)-93
Abstract
Background: Platelet-rich plasma (PRP) is prepared with centrifugation protocols that differ by more than an order of magnitude in relative centrifugal force and by a factor of eight in spin time. It is widely assumed that these differences drive platelet concentration, platelet integrity, growth factor content and, ultimately, clinical benefit. That chain of assumptions has never been tested end to end against the whole published record.
Objective: To quantify, in a single integrated analysis, (i) how centrifugation parameters relate to platelet concentration factor and platelet recovery, (ii) what evidence exists that centrifugation lyses or activates platelets, (iii) how growth factor content varies across protocols, and (iv) whether any of this propagates into clinical outcomes across the main regenerative medicine indications.
Methods: Three parallel evidence streams were assembled and analysed. Stream 1: 59 laboratory preparation arms from human whole-blood centrifugation studies retrieved from 40 distinct source documents, plus a 34-row commercial device series. Stream 2: 57 randomised controlled trials of PRP across seven indications, with arm-level data extracted where reported. Stream 3: an umbrella synthesis of 190 pooled estimates from 36 systematic reviews and meta-analyses. Concentration factor, recovery, growth factor concentrations and effect sizes were pooled with descriptive and random-effects methods; the relationship between centrifugation parameters and platelet yield was tested by weighted meta-regression on the log scale; between-group differences used Hedges g with Mann–Whitney confirmation.
Results: The pooled platelet concentration factor was 4.63× (SD 2.22; median 4.53×; range 1.47–9.50×; k = 42), higher than the 3–5× figure usually quoted. Double-spin protocols reached 4.96× versus 2.81× for single-spin protocols (difference 2.15×, 95% CI 0.91–3.38; Hedges g = 1.20; p = 0.0039), and leukocyte-rich preparations reached 5.90× versus 3.27× for leukocyte-poor preparations (p = 0.0004). Critically, neither first-spin g-force (β = -0.042, R² = 0.006, p = 0.71) nor second-spin g-force (β = -0.011, p = 0.94) predicted concentration factor; only cumulative centrifugal dose (g·min) showed a weak positive trend (β = 0.221, R² = 0.133, p = 0.079). Platelet recovery averaged 68.4% (range 21.8–92.0%). Direct integrity markers were discordant: several groups reported activation or viability loss above 800–3,000 g, whereas others found no change in P-selectin, lactate dehydrogenase or activated-platelet fraction. Growth factor concentrations spanned two to three orders of magnitude between studies and did not correlate with concentration factor (all p > 0.10). In the clinical streams, PRP outperformed comparators in 120 of 190 pooled estimates, but only 16 of 57 randomised trials (28%) reported a full preparation protocol and only 7 reported the delivered platelet concentration factor. Platelet dose — not leukocyte class — was the only composition variable with a consistent signal.
Conclusions: Centrifugation reliably determines what enters the syringe: spin architecture (single versus double) and cumulative centrifugal dose govern platelet yield, while the raw g-force of any individual spin does not. Evidence that clinically used protocols lyse platelets is weak and contradicted by direct integrity assays. The link from preparation to patient outcome is broken not by biology but by reporting: the great majority of randomised trials do not state what they injected. Mandatory reporting of platelet dose is the single highest-yield change available to the field.
Keywords
Platelet-rich plasma; Centrifugation; Platelet concentration factor; Platelet lysis; Growth factors; Meta-analysis; Regenerative medicine; Osteoarthritis; Tendinopathy.
Introduction
Platelet-rich plasma occupies an unusual position in regenerative medicine. It is autologous, inexpensive relative to cell therapies, and prepared at the point of care in minutes, yet after three decades of use the field still disagrees about whether it works, for whom, and — most consequentially — about what the product actually is. The core difficulty is that “PRP” does not name a substance. It names the output of a procedure, and that procedure varies enormously.
Published human whole-blood protocols in this analysis span first-spin relative centrifugal forces from 44 g to 3,000 g and spin durations from 3 to 30 minutes, with second spins from 130 g to 3,000 g. The same 20 mL of venous blood, processed by two protocols that are both described in print as “standard PRP”, can yield a platelet concentration factor of 1.47× or 9.50×, leukocyte counts that differ by more than an order of magnitude, and growth factor concentrations that differ by two. A field with that much variance in its independent variable cannot produce stable conclusions about its dependent variable.
Three mechanistic claims dominate the practical literature and shape how clinicians choose kits. The first is that higher centrifugal force concentrates platelets more effectively, up to a limit. The second is that beyond some threshold — most often quoted as 3,000 g — centrifugation damages, activates or lyses platelets, degranulating them prematurely and wasting the growth factor payload. The third is that growth factor content scales with platelet count, so that a higher concentration factor delivers a proportionally stronger biological stimulus. Each claim is plausible. Each is repeated in review after review. None has been tested against the assembled quantitative record.
The clinical literature has developed largely in parallel and largely indifferent to this problem. Randomised trials compare “PRP” with saline, corticosteroid or hyaluronic acid, and meta-analyses pool them. When those meta-analyses find heterogeneity — and they almost always do — preparation differences are named as the likely culprit in the discussion section. But if the trials themselves do not report their preparation, that explanation can never be tested; it functions as an unfalsifiable placeholder rather than a hypothesis.
This study was designed to close the loop. Rather than analysing preparation science or clinical outcomes in isolation, it assembles three evidence streams — laboratory centrifugation studies, randomised controlled trials, and the existing meta-analytic literature — and asks a single connected question: does the way PRP is spun change what is in the syringe, and does what is in the syringe change what happens to the patient? The two halves of that question, as will be shown, have very different answers.
Methods
Design and reporting
This is a systematic quantitative synthesis of three parallel evidence streams, reported in accordance with PRISMA 2020 for the trial and review streams and with descriptive meta-analytic conventions for the laboratory stream. Because the laboratory stream pools preparation arms rather than randomised comparisons, it is analysed with descriptive pooling, between-group testing and meta-regression rather than with inverse-variance effect-size pooling. No ethical approval was required; all data derive from published sources.
Eligibility criteria
Stream 1 (laboratory) included studies performing centrifugation of human whole blood and reporting at least one of: platelet concentration factor, absolute platelet counts before and after processing, platelet recovery or efficiency, leukocyte content, growth factor concentrations, or direct markers of platelet activation, viability or lysis. Animal studies were excluded; a rabbit-blood study otherwise meeting criteria was excluded on this basis. Narrative reviews without primary measurements were excluded from the data rows but retained for contextual citation.
Stream 2 (randomised trials) included randomised controlled trials of PRP in human musculoskeletal, cutaneous or dermatological indications with a non-PRP or alternative-PRP comparator and a validated outcome measure. Stream 3 (umbrella) included systematic reviews with quantitative pooling of PRP outcomes; each pooled estimate was captured as a separate row, annotated with the review, indication, comparison, outcome, timepoint, number of contributing trials, metric, point estimate, confidence interval and reported I².
Information sources and extraction
Sources were retrieved as full text where openly available and as structured abstract records otherwise. Every numeric value in this manuscript is bound to the document from which it was read; values that could not be confirmed from a retrieved page were recorded as missing rather than imputed. The laboratory stream comprises 59 preparation arms drawn from 40 distinct source documents (several studies contribute multiple arms because they tested multiple speeds, times or devices). The clinical stream comprises 63 extracted outcome rows from 57 unique trials. The umbrella stream comprises 190 pooled estimates from 36 reviews.
Unit harmonisation was applied before analysis: platelet counts expressed as ×10⁹/L, ×10³/µL and ×10⁶/mL were treated as equivalent; counts in cells/mm³ and /µL were divided by 1,000; growth factors in ng/mL were multiplied by 1,000 to pg/mL. Where a study reported only a range, the range was retained descriptively and the analytic cell left missing — no midpoints were invented. Where a study reported both counts but not the ratio, the concentration factor was computed as PRP platelet count divided by whole-blood platelet count from the same page.
Statistical analysis
Concentration factor, recovery and growth factor concentrations were summarised as mean, standard deviation, median, interquartile range and range. Differences between spin architectures (single versus double) and between leukocyte classes (leukocyte-poor versus leukocyte-rich) were tested with Welch t-tests and confirmed with Mann–Whitney U tests, given the modest group sizes and right-skewed distributions; effect sizes are reported as Hedges g with 95% confidence intervals for the raw difference. The relationship between centrifugation parameters and yield was modelled as ordinary least squares regression of natural-log concentration factor on natural-log first-spin g, natural-log second-spin g, and natural-log cumulative centrifugal dose (the product of g and minutes summed across spins), with slope, R² and p reported.
For clinical trials reporting arm-level means, standard deviations and sample sizes, Hedges g was computed with small-sample correction and pooled under a DerSimonian–Laird random-effects model, with τ², Cochran Q and I². Scales on which higher scores denote better function (IKDC, Constant–Murley, AOFAS, hair density) were sign-reversed so that a positive g always favours PRP. Two trials reporting standard errors in place of standard deviations were excluded from the primary pool and retained only in a sensitivity analysis, since treating standard errors as standard deviations inflates effect sizes. Umbrella estimates were not re-pooled across reviews, because reviews overlap substantially in their constituent trials; they are reported as a distribution of pooled estimates with their reported heterogeneity. Analyses were performed in Python 3 with NumPy and SciPy; two-sided α = 0.05.
Risk of bias and certainty
For the trial stream, blinding status was extracted as an index of internal validity, and completeness of preparation reporting was extracted as an index of reproducibility — a domain not covered by standard risk-of-bias tools but decisive for this research question. For the umbrella stream, reported I² was used as the primary indicator of consistency. Because the laboratory stream is composed of non-randomised descriptive arms, no formal risk-of-bias instrument was applied; instead each arm carries its source binding and analytic decisions are stated explicitly in the notes to each table.
Results
Evidence base
Figure 1: Three-stream evidence flow. Laboratory preparation arms, randomised controlled trials and previously published pooled estimates were assembled in parallel and analysed with stream-appropriate methods. Counts shown are the records retained for analysis after unit harmonisation and exclusion of non-human and non-quantitative sources.
The final evidence base comprised 59 laboratory preparation arms from 40 source documents, 57 randomised controlled trials contributing 63 outcome rows, and 190 pooled estimates from 36 systematic reviews. Indication coverage in the trial stream was dominated by knee osteoarthritis (27 trials), followed by lateral epicondylitis (7), rotator cuff pathology (7 across repair, tendinopathy and partial tear), Achilles tendinopathy (4), plantar fasciitis (4), chronic and diabetic wounds (5) and androgenetic alopecia (3).
Laboratory characteristics
(Table 1) lists every laboratory preparation arm with its spin architecture, centrifugation parameters, platelet counts, concentration factor, recovery and leukocyte class. (Table 2) summarises the commercial device series, which is analysed separately because device manufacturers frequently do not disclose internal spin parameters and because several device rows derive from a single standardised reanalysis rather than from independent laboratories.
|
Study arm |
n |
Design |
Spin 1 (g / min) |
Spin 2 (g / min) |
WB PLT |
PRP PLT |
Fold |
Rec. % |
Leuko |
|
Perez 2013 |
10 |
single |
100 / 10 |
n.a. |
245 |
467 |
1.9 |
70.8 |
poor |
|
Perez 2014 |
20 |
double |
100 / 10 |
400 / 10 |
250 |
668 |
3.1 |
n.a. |
poor |
|
Amable 2013 |
22 |
double |
300 / 5 |
700 / 17 |
n.a. |
n.a. |
n.a. |
n.a. |
poor |
|
Bausset 2012 |
54 |
double |
130 / 15 |
130 / 15 |
260 |
n.a. |
3.01 |
n.a. |
poor |
|
Bausset 2012 250g |
54 |
double |
130 / 15 |
250 / 15 |
260 |
n.a. |
3.47 |
n.a. |
poor |
|
Bausset 2012 400g |
54 |
double |
130 / 15 |
400 / 15 |
260 |
n.a. |
3.73 |
n.a. |
poor |
|
Bausset 2012 1000g |
54 |
double |
130 / 15 |
1,000 / 15 |
260 |
n.a. |
3.96 |
n.a. |
poor |
|
Fitzpatrick 2017 GPSIII |
3 |
single |
1,100 / 15 |
n.a. |
269 |
964 |
3.58 |
n.a. |
rich |
|
Fitzpatrick 2017 SmartPrep2 |
3 |
single |
1,250 / 14 |
n.a. |
269 |
1,224.00 |
4.55 |
n.a. |
rich |
|
Fitzpatrick 2017 Magellan |
3 |
single |
1,200 / 17 |
n.a. |
269 |
1,266.00 |
4.71 |
n.a. |
rich |
|
Fitzpatrick 2017 ACP |
3 |
single |
1,500 / 5 |
n.a. |
269 |
412 |
1.53 |
n.a. |
poor |
|
Magalon 2016 DEPA |
20 |
both |
n.a. |
n.a. |
200 |
n.a. |
n.a. |
n.a. |
both |
|
Piao 2017 |
n.a. |
double |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
89 |
n.a. |
|
Yin 2016 PPRP |
10 |
double |
160 / 10 |
250 / 15 |
228.3 |
1,461.80 |
6.4 |
n.a. |
poor |
|
Yin 2016 LPRP |
10 |
double |
160 / 10 |
250 / 15 |
228.3 |
1,436.70 |
6.29 |
n.a. |
rich |
|
Roh 2016 SS |
14 |
single |
900 / 5 |
n.a. |
147 |
311 |
2.12 |
92 |
poor |
|
Roh 2016 DS |
14 |
double |
900 / 5 |
1,500 / 15 |
147 |
1,145.00 |
7.79 |
84.3 |
rich |
|
Etulain 2018 |
8 |
single |
180 / 10 |
890 / 10 |
168 |
453 |
2.7 |
n.a. |
n.a. |
|
Taniguchi 2019 |
39 |
double |
580 / 8 |
1,000 / 20 |
234 |
414 |
1.8 |
n.a. |
poor |
|
Dejnek 2022 ACP |
12 |
single |
n.a. / 5 |
n.a. |
240.7 |
357.3 |
1.47 |
43.7 |
poor |
|
Dejnek 2022 MiniGPSIII |
12 |
single |
n.a. / 15 |
n.a. |
240.7 |
1,212.70 |
5.05 |
56.1 |
rich |
|
Dejnek 2022 Xerthra |
12 |
single |
n.a. / 5 |
n.a. |
240.7 |
455.3 |
1.96 |
21.8 |
poor |
|
Dejnek 2022 DrPRP |
12 |
single |
n.a. / 4 |
n.a. |
240.7 |
499.8 |
2.14 |
35.6 |
poor |
|
Muthu 2022 |
40 |
double |
100 / 15 |
1,600 / 20 |
179.5 |
n.a. |
6.4 |
n.a. |
n.a. |
|
Machado 2022 |
16 |
double |
200 / 12 |
1,600 / 8 |
251.2 |
847.2 |
3.47 |
n.a. |
rich |
|
Carvalho 2024 |
21 |
double |
300 / 10 |
700 / 15 |
200.2 |
1,858.40 |
9.28 |
29.9 |
n.a. |
|
Husak 2025 PRP |
30 |
double |
600 / 10 |
1,600 / 7 |
236 |
987 |
4.1 |
78.7 |
poor |
|
Husak 2025 LPRP |
30 |
double |
600 / 10 |
1,600 / 7 |
236 |
n.a. |
4.9 |
89.4 |
rich |
|
Miroshnychenko 2020 |
2 |
n.a. |
n.a. |
n.a. |
152 |
685 |
4.5 |
n.a. |
poor |
|
Huber 2016 |
9 |
double |
300 / 5 |
700 / 17 |
270.4 |
1,912.00 |
7.1 |
n.a. |
rich |
|
Kruel 2026 |
34 |
double |
300 / 5 |
700 / 17 |
231.7 |
1,416.80 |
6.11 |
n.a. |
rich |
|
Garrison 2026 Angel |
30 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
6.31 |
n.a. |
n.a. |
|
Garrison 2026 Magellan |
30 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
6.58 |
n.a. |
n.a. |
|
Panero 2025 AngelLP |
3 |
n.a. |
n.a. |
n.a. |
n.a. |
1,221.00 |
5.7 |
n.a. |
poor |
|
Panero 2025 AngelLR |
3 |
n.a. |
n.a. |
n.a. |
n.a. |
1,756.00 |
8.8 |
n.a. |
rich |
|
Panero 2025 ApexLP |
3 |
n.a. |
n.a. / 10 |
n.a. |
n.a. |
1,334.00 |
6.6 |
n.a. |
poor |
|
Panero 2025 ApexLR |
3 |
n.a. |
n.a. / 10 |
n.a. |
n.a. |
1,822.00 |
9.5 |
n.a. |
rich |
|
Okumo 2023 |
9 |
n.a. |
n.a. |
n.a. |
238 |
1,434.00 |
6.03 |
n.a. |
rich |
|
Menon 2026 |
10 |
single |
44 / 8 |
n.a. |
n.a. |
578 |
n.a. |
n.a. |
poor |
|
Araki 2012 |
9 |
single |
250 / 10 |
2,330 / 10 |
245 |
627 |
2.56 |
88.7 |
poor |
|
Mazzocca 2012 |
8 |
both |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
|
Weibrich 2002 |
213 |
n.a. |
n.a. |
n.a. |
266 |
1,407.60 |
5 |
n.a. |
n.a. |
|
Eppley 2004 |
10 |
n.a. |
n.a. |
n.a. |
197 |
1,600.00 |
8 |
n.a. |
n.a. |
|
Sundman 2011 ACP |
11 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
1.99 |
n.a. |
poor |
|
Sundman 2011 GPSIIIMini |
11 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
4.69 |
n.a. |
rich |
|
Jo 2013 |
39 |
double |
900 / 5 |
1,500 / 15 |
n.a. |
n.a. |
n.a. |
92 |
n.a. |
|
Slichter 1976 |
n.a. |
double |
1,000 / 9 |
3,000 / 20 |
n.a. |
n.a. |
n.a. |
86 |
n.a. |
|
Tamimi 2007 ACE |
30 |
double |
n.a. |
n.a. |
n.a. |
n.a. |
3.36 |
n.a. |
n.a. |
|
Tamimi 2007 Nahita |
30 |
single |
n.a. |
n.a. |
n.a. |
n.a. |
2.27 |
n.a. |
n.a. |
|
Oh 2015 SS |
14 |
single |
900 / 5 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
poor |
|
Oh 2015 DS |
14 |
double |
900 / 5 |
1,500 / 15 |
n.a. |
n.a. |
n.a. |
n.a. |
rich |
|
Kobayashi 2016 |
6 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
both |
|
Magalon 2014 |
10 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
both |
|
Weibrich 2012 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
|
Ozer 2019 |
n.a. |
single |
1,630 / 5 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
|
Xie 2025 |
15 |
double |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
|
Croise 2020 |
n.a. |
both |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
|
DohanEhrenfest 2009 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
both |
|
Landesberg 2000 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
Table 1: Laboratory preparation arms (k = 59 from 40 source documents). WB PLT and PRP PLT in ×10³/µL after unit harmonisation. Fold = platelet concentration factor. Rec. % = platelet recovery or efficiency as printed or as computed from printed counts. Leuko = leukocyte class (poor / rich) as designated by the source. Several studies contribute multiple arms where they tested more than one speed, time or device.
Missing cells indicate the value was not reported on the retrieved page; no values were imputed.
|
Device |
Source |
Spin protocol |
WB PLT |
PRP PLT |
Fold |
Recovery / efficiency % |
Leukocytes |
|
GPS III |
Fitzpatrick 2017 |
1100 g / 15 min |
269 ± 106 |
964 ± 551 |
3.58 calc |
n.a. |
LR, 35.8 ± 10.8 ×10³/µL |
|
GPS III |
Magalon 2016 DEPA (Castillo data) |
single 1100 g / 15 min |
n.a. |
n.a. |
n.a. |
22.6 (poor); purity 27.0%; dose 2.48 billion |
whole-blood-type PRP |
|
Mini GPS III |
Magalon 2016 DEPA |
single 3200 rpm / 15 min |
n.a. |
n.a. |
n.a. |
34.6; purity 51.8%; dose 2.56 billion |
heterogeneous |
|
Mini GPS III |
Dejnek 2022 |
single 3200 rpm / 15 min |
240.67 ± 49.85 |
1212.67 ± 268.63 |
5.05 ± 0.67 |
56.15 ± 7.44 |
LR, 34.19 ± 11.18 ×10⁹/L |
|
GPS III Mini |
Sundman 2011 |
n.a. |
n.a. |
n.a. |
4.69 |
n.a. |
LR, 4.26× blood |
|
GPS II |
Magalon 2016 DEPA (Kaux data) |
single 180 g / 15 min |
n.a. |
n.a. |
n.a. |
22.8; purity 6.0%; dose 2.28 billion |
whole-blood-type PRP |
|
Angel (Arthrex) |
Garrison 2026 |
n.a. |
95.4–256.2 |
537.5–1438 |
6.31 ± 0.93 |
n.a. |
n.a. |
|
Angel LP / LR |
Panero 2025 |
2% / 7% haematocrit settings |
n.a. |
1221 ± 955 / 1756 ± 312 ×10⁶/mL |
5.7 ± 4.2 / 8.8 ± 0.2 |
n.a. |
10.0 ± 2.2 / 25.9 ± 7.3 ×10⁶/mL |
|
Magellan |
Fitzpatrick 2017 |
1200 g / 17 min |
269 ± 106 |
1266 ± 831 |
4.71 calc |
n.a. |
LR, 31.4 ± 9.4 ×10³/µL |
|
Magellan |
Garrison 2026 |
n.a. |
69.1–245.3 |
478.5–1343.1 |
6.58 ± 1.33 |
n.a. |
n.a. |
|
Magellan |
Magalon 2016 DEPA (Castillo) |
single 1200 g / 17 min |
n.a. |
n.a. |
n.a. |
65.8; purity 60.4%; dose 3.41 billion |
heterogeneous |
|
Magellan |
Magalon 2016 DEPA (Kushida) |
double 610 g/4 min + 1240 g/6 min |
n.a. |
n.a. |
n.a. |
45.3; purity 32.9%; dose 5.43 billion (grade A) |
heterogeneous |
|
ACP (Arthrex) |
Fitzpatrick 2017 |
1500 g / 5 min |
269 ± 106 |
412 ± 140 |
1.53 calc |
n.a. |
LP, 1.3 ± 0.781 ×10³/µL |
|
ACP (Arthrex) |
Dejnek 2022 |
1500 rpm / 5 min |
240.67 ± 49.85 |
357.33 ± 99.01 |
1.47 ± 0.18 |
43.68 ± 5.32 |
LP, 0.87 ± 1.01 ×10⁹/L |
|
ACP (Arthrex) |
Sundman 2011 |
n.a. |
n.a. |
n.a. |
1.99 |
n.a. |
LP, 0.13× blood |
|
Arthrex (ACP) |
Magalon 2016 DEPA |
single 1500 rpm / 5 min |
n.a. |
n.a. |
n.a. |
48.0; purity 81.0%; dose 1.06 billion |
pure PRP |
|
SmartPrep2 (Harvest Terumo) |
Fitzpatrick 2017 |
1250/1050 g / 14 min |
269 ± 106 |
1224 ± 560 |
4.55 calc |
n.a. |
LR, 24.7 ± 8.69 ×10³/µL |
|
RegenLab / RegenPRP |
Magalon 2016 DEPA |
single 300 g / 5 min (Kaux); single 1500 g / 9 min (Magalon) |
n.a. |
n.a. |
n.a. |
79.3 (purity 97.5%, dose 0.95 bn) / 61.7 (purity 46.0%, dose 0.99 bn) |
very pure / heterogeneous |
|
Cascade |
Magalon 2016 DEPA (Castillo) |
single 1100 g / 6 min |
n.a. |
n.a. |
n.a. |
67.5; purity 81.5%; dose 2.43 billion |
pure PRP |
|
Selphyl |
Magalon 2016 DEPA |
single 1100 g / 6 min (Magalon); single 525 g / 15 min (Kushida) |
n.a. |
n.a. |
n.a. |
59.5 (purity 73.9%, dose 0.95 bn) / 13.1 (purity 99.7%, dose 0.21 bn) |
pure / very pure |
|
Curasan |
Magalon 2016 DEPA (Kaux) |
double 1000 g/10 min + 2300 g/15 min |
n.a. |
n.a. |
n.a. |
32.4; purity 97.7%; dose 0.55 billion |
very pure |
|
Plateltex |
Magalon 2016 DEPA (Kaux) |
double 180 g/10 min + 1000 g/10 min |
n.a. |
n.a. |
3.43 |
19.4; purity 87.5%; dose 0.23 billion |
pure |
|
JP200 |
Magalon 2016 DEPA (Kushida) |
double 1000 g/6 min + 800 g/8 min |
n.a. |
n.a. |
n.a. |
26.0; purity 19.6%; dose 1.04 billion |
whole-blood-type |
|
GLO PRP |
Magalon 2016 DEPA (Kushida) |
double 1800 g/3 min + 1800 g/6 min |
n.a. |
n.a. |
n.a. |
37.4; purity 38.2%; dose 0.64 billion |
heterogeneous |
|
Kyocera |
Magalon 2016 DEPA (Kushida) |
double 600 g/7 min + 2000 g/5 min |
n.a. |
n.a. |
n.a. |
78.1; purity 29.4%; dose 3.12 billion |
whole-blood-type |
|
MyCells |
Magalon 2016 DEPA (Kushida) |
single 2054 g / 7 min |
n.a. |
n.a. |
n.a. |
48.8; purity 87.3%; dose 0.98 billion |
pure |
|
Dr. Shin's THROMBO KIT |
Magalon 2016 DEPA (Kushida) |
single 1720 g / 8 min |
n.a. |
n.a. |
n.a. |
45.9; purity 18.8%; dose 0.78 billion |
whole-blood-type |
|
Xerthra |
Dejnek 2022 |
3500 rpm / 5 min |
240.67 ± 49.85 |
455.27 ± 362.92 |
1.96 ± 1.71 |
21.79 ± 18.98 |
LP, 1.80 ± 2.55 ×10⁹/L |
|
Dr. PRP |
Dejnek 2022 |
3100 rpm / 4 min |
240.67 ± 49.85 |
499.75 ± 153.46 |
2.14 ± 0.73 |
35.61 ± 12.13 |
LP, 0.60 ± 0.87 ×10⁹/L |
|
Apex Biologics / XCELL LP & LR |
Panero 2025 |
3800 rpm / 10 min |
n.a. |
1334 ± 838 / 1822 ± 256 ×10⁶/mL |
6.6 ± 3.1 / 9.5 ± 1.3 |
n.a. |
2.9 ± 2.1 / 29.0 ± 8.6 ×10⁶/mL |
|
PEAK (DePuy Synthes Mitek) |
Okumo 2023 |
n.a. |
238 ± 44 |
1434 ± 391 |
6.03 calc |
n.a. |
LR, 23,006 ± 6819/µL |
|
Biomet 3i Platelet Concentrate Collection System |
Weibrich 2012 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
highest growth factor levels of the 6 methods |
|
ACE (double) vs Nahita (single) |
Tamimi 2007 |
n.a. |
n.a. |
n.a. |
336% / 227% |
n.a. |
n.a. |
|
Prodizen Prosys |
Oh 2015 |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
n.a. |
Table 2: Commercial PRP systems (k = 34 device rows). Counts in ×10³/µL unless otherwise stated. Purity and absolute platelet dose values are reproduced as printed in the source reanalysis. LR = leukocyte-rich; LP = leukocyte-poor.
Device rows attributed to a standardised reanalysis are labelled with the originating dataset in parentheses.
Pooled platelet concentration and the effect of spin architecture
Across 42 arms reporting a concentration factor, the pooled mean was 4.63× (SD 2.22), median 4.53×, interquartile range 2.78–6.30× and full range 1.47–9.50×. This is materially higher than the 3–5× figure conventionally cited, and the interquartile range alone spans a 2.3-fold difference in delivered platelet dose between the 25th and 75th percentile protocol.
Spin architecture was the dominant determinant. Double-spin protocols achieved a mean concentration factor of 4.96× (SD 2.03, k = 17) against 2.81× (SD 1.24, k = 13) for single-spin protocols — an absolute difference of 2.15× (95% CI 0.91–3.38), Hedges g = 1.20, Welch t = -3.66, p = 0.0011, Mann–Whitney p = 0.0039. Leukocyte-rich preparations reached 5.90× (k = 14) versus 3.27× (k = 19) for leukocyte-poor preparations, a difference of 2.63× (p = 0.0004) that reflects the buffy-coat harvesting strategy common to leukocyte-rich systems rather than an independent effect of leukocytes themselves.
|
Stratum |
k arms |
Mean fold |
SD |
Median |
Range |
Test statistic |
p |
|
All arms |
42 |
4.63 |
2.22 |
4.53 |
1.47–9.50 |
— |
— |
|
Single spin |
13 |
2.81 |
1.24 |
2.27 |
1.47–5.05 |
Welch t = -3.66 |
0.0011 |
|
Double spin |
17 |
4.96 |
2.03 |
4.1 |
1.80–9.28 |
Mann–Whitney U = 41.0 |
0.0039 |
|
Leukocyte-poor |
19 |
3.27 |
1.61 |
3.01 |
— |
Mann–Whitney |
0.0004 |
|
Leukocyte-rich |
14 |
5.9 |
1.85 |
5.54 |
— |
— |
— |
|
Platelet recovery (%) |
14 |
68.4 |
25.7 |
81.5 |
21.8–92.0 |
— |
— |
Table 3: Pooled platelet metrics by spin architecture and leukocyte class. Hedges g for the single- versus double-spin contrast was 1.20 with an absolute difference of 2.15× (95% CI 0.91–3.38). Recovery is reported on a different scale (percentage of whole-blood platelets captured) and is shown here for completeness.
Figure 2: Platelet concentration factor by spin architecture and leukocyte class. Boxes show median and interquartile range, whiskers the full range, and individual points each preparation arm. Double-spin and leukocyte-rich protocols concentrate platelets more effectively; the two strata overlap substantially because buffy-coat systems are usually both.
Centrifugal force does not predict platelet yield
The central mechanical assumption of the field — that the g-force chosen determines how many platelets are recovered — was not supported. Regressing log concentration factor on log first-spin g across 24 arms produced a slope of -0.042 (SE 0.112) with R² = 0.006 and p = 0.71: statistically indistinguishable from no relationship, and explaining well under one percent of the variance. The second spin behaved the same way (18 arms, slope -0.011, R² = 0.000, p = 0.94). Only cumulative centrifugal dose — g multiplied by minutes, summed across spins — showed a positive trend, and even that fell short of conventional significance (24 arms, slope 0.221, R² = 0.133, p = 0.079), accounting for roughly 13% of the variance.
Read together with the strong architecture effect in (Table 3), this has a clear interpretation. What matters is not how hard the blood is spun but how many separation steps are performed and how much total work is done. A protocol that separates plasma from red cells and then re-concentrates the platelet fraction will outperform a single-step protocol almost regardless of the speeds selected within the range in clinical use.
|
Predictor |
k arms |
β (log-log slope) |
SE |
R² |
p |
Interpretation |
|
First-spin g-force |
24 |
-0.042 |
0.112 |
0.006 |
0.711 |
No relationship |
|
Second-spin g-force |
18 |
-0.011 |
0.135 |
0 |
0.938 |
No relationship |
|
Cumulative dose (g·min) |
24 |
0.221 |
0.12 |
0.133 |
0.079 |
Weak positive trend |
|
Spin architecture (single vs double) |
30 |
Δ = 2.15× |
— |
g = 1.20 |
0.0039 |
Strong effect |
|
Leukocyte class (poor vs rich) |
33 |
Δ = 2.63× |
— |
— |
0.0004 |
Strong effect |
Table 4: Determinants of platelet concentration factor. Regressions are ordinary least squares on natural-log transformed variables; architecture and leukocyte contrasts are between-group comparisons with Mann–Whitney confirmation. The two variables’ clinicians most often adjust — the g-force of each spin — are precisely the two with no measurable effect.
Figure 3: Concentration factor against centrifugation parameters. Panels show first-spin g, second-spin g and cumulative centrifugal dose on log axes with fitted regression lines. The flat fits in the first two panels are the central negative finding of the laboratory stream; the modest positive slope in the third panel indicates that total work, not peak force, is the operative variable.
Platelet recovery and the shape of the loss curve
Recovery — the proportion of whole-blood platelets captured in the final product — averaged 68.4% (SD 25.7, median 81.5%, range 21.8–92.0%, k = 14). The spread is the point: a protocol at the bottom of this distribution discards four platelets for everyone it delivers. Unlike concentration factor, recovery does show force dependence in the individual studies that manipulated force systematically, but the direction is not monotonic and differs between the first and second spin.
Several groups reported first-spin recovery peaking in the 70–100 g range and falling steeply thereafter, with losses of 30–40% attributed to rapid, dense packing of erythrocytes that traps platelets in the cell pellet. Others located the first-spin optimum higher, at 190–320 g, with progressive decline to 2,330 g. For the second spin the pattern reverses: recovery has been reported to rise monotonically with force from 69.0% at 1,010 g to 91.2% at 2,330 g, because at that stage the objective is to pellet a platelet-only suspension rather than to avoid co-pelleting with red cells. This is why a single “optimal g-force” cannot exist: the two spins have opposite physics.
Figure 4: Platelet recovery against first-spin g-force. Points are individual preparation arms; shaded bands mark the force ranges around which published optima and decline thresholds cluster. The dispersion within every band shows that force alone does not determine recovery.
Does centrifugation destroy platelets?
The claim that centrifugation above a threshold lyses or exhausts platelets is the most consequential mechanistic assertion in the field, because it is the stated reason for avoiding high-speed protocols. Table 5 collects the primary statements, quoted from their sources, together with the direct integrity measurements that accompany them.
|
Source |
Domain |
Reported threshold or finding (as printed) |
Evidence type |
|
Perez 2013 |
Recovery |
Recovery rose to ~70% up to 100 g then fell sharply to 8% approaching 820 g; dense red-cell packing caused platelet losses of 30–40% |
Indirect (recovery) |
|
Perez 2014 |
Activation |
Soluble P-selectin increased at 800 g and 1,200 g, interpreted as loss of platelet integrity; recovery peaked at 70–100 g |
Direct (sP-selectin) |
|
Bausset 2012 |
Activation |
Second-spin centrifugation below 400 g maintained procoagulant activity; 1,000 g induced aggregation, reduced ADP-induced P-selectin expression and morphological change |
Direct (flow cytometry, morphology) |
|
Piao 2017 |
Viability |
“Platelet viability is reduced significantly when acceleration becomes greater than 3,000 g, applied for 20 minutes”; leukocyte recovery near zero above ~800 g |
Direct (viability) |
|
Amable 2013 |
Yield/recovery |
Increasing force from 200 to 360 g was detrimental to both yield and recovery, as was increasing time |
Indirect (yield) |
|
Araki 2012 |
Recovery |
First-spin collection peaked at 190–320 g and declined to 2,330 g; second-spin recovery instead rose from 69.0% at 1,010 g to 91.2% at 2,330 g |
Indirect (recovery) |
|
Muthu 2022 |
Integrity |
First-spin recovery peaked at 100 g and declined from 600 g; second-spin recovery peaked at 1,600 g; lactate dehydrogenase 213 ± 24 → 237 ± 38 U/L, p = 0.432 — no significant damage at 1,600 g for 20 min |
Direct (LDH) — negative |
|
Eppley 2004 |
Activation |
P-selectin unchanged during concentration (45 ± 16 → 52 ± 11 pg/mL, p = 0.65) |
Direct (P-selectin) — negative |
|
Magalon 2014 |
Activation |
No difference in the percentage of activated platelets between centrifugation protocols |
Direct — negative |
|
Xie 2025 |
Activation |
No significant CD62P difference between fixed-angle and horizontal rotors |
Direct (CD62P) — negative |
|
Ozer 2019 |
Composite |
No significant differences across 45–1,630 g and 5–20 min (p > 0.05); recommends trading force against time |
Composite — negative |
|
Slichter & Harker 1976 |
Viability |
1,000 g for 9 min followed by 3,000 g for 20 min harvested 86 ± 1% of platelets “without loss of viability” |
Direct — negative at 3,000 g |
|
Tamimi 2007 |
Ultrastructure |
Transmission electron microscopy showed aggregate alteration in a double-spin system |
Direct (TEM) — positive |
|
Croise 2020 |
Narrative |
“When the centrifugation force is extended, platelets can possibly be altered” |
Review statement |
Table 5: Published centrifugation damage and recovery thresholds, with the type of evidence supporting each. Statements are reproduced from the source text. Note the asymmetry: the widely quoted 3,000 g lysis threshold rests on a single viability study, while four independent direct-integrity assays — P-selectin, CD62P, lactate dehydrogenase and activated-platelet fraction — found no damage at forces in routine clinical use.
The balance of evidence does not support a simple lysis threshold at clinically used forces. What the negative studies share is that they measured platelet integrity directly; what several of the positive studies share is that they inferred damage from falling recovery, which is equally consistent with platelets being trapped in the red-cell pellet rather than destroyed. A trapped platelet and a lysed platelet look identical in a recovery calculation and completely different under a flow cytometer. The one report of frank viability loss above 3,000 g is not contradicted here — but 3,000 g for 20 minutes lies at or beyond the upper edge of routine practice, and the historical haematology literature reports 3,000 g for 20 minutes harvesting 86% of platelets without loss of viability.
Growth factor content is not predicted by platelet count
|
Growth factor |
k arms |
Median (pg/mL) |
Mean (pg/mL) |
Range (pg/mL) |
Fold spread across studies |
|
PDGF-BB |
6 |
8,395.00 |
26,874.70 |
342.0–126,186.0 |
369× |
|
TGF-β1 |
9 |
89,000.00 |
76,047.80 |
750.0–169,000.0 |
225× |
|
VEGF |
9 |
157.9 |
590.5 |
4.2–1,837.0 |
438× |
|
EGF |
2 |
534.6 |
534.6 |
470.0–599.3 |
1× |
Table 6: Growth factor concentrations in PRP across laboratory arms. Values harmonised to pg/mL. The final column is the ratio of the highest to the lowest reported value and is the key number: the same named growth factor in the same named product varies by two to three orders of magnitude between laboratories.
Studies reporting growth factors per 10⁶ platelets rather than per volume are excluded from this table and reported descriptively in the source extraction.
Median concentrations were 8,395 pg/mL for PDGF-BB (k = 6), 89,000 pg/mL for TGF-β1 (k = 9), 158 pg/mL for VEGF (k = 9) and 535 pg/mL for EGF (k = 2). Correlations between concentration factor and growth factor concentration were positive in direction but none reached significance, the strongest being TGF-β1 (r = 0.61, p = 0.11). Assay platform, activation state at the time of measurement, and whether the sample was lysed before assay plausibly contribute more variance than the platelet count itself — which is precisely why platelet dose, not growth factor concentration, is the practical reporting target.
Figure 5: Growth factor concentrations by study. Lollipop plots on a logarithmic scale for PDGF-BB, TGF-β1 and VEGF. Each point is a single preparation arm. The horizontal spread within each panel — not the differences between panels — is the finding.
The reporting gap in randomised trials
Of 57 randomised trials, 16 (28%) reported a full preparation protocol, 18 (32%) reported partial detail, and 23 (40%) reported none. Only 7 trials stated the platelet concentration factor actually delivered, and those that did span 1.99× to 5.68× — a range across which the laboratory stream predicts a nearly fourfold difference in platelet dose. Blinding was better reported: 32 trials were double-blind, 6 single-blind, 5 open-label and 14 unclear.
|
Trial |
Indication |
Comparator |
Preparation reporting |
Stated fold |
Blinding |
Outcome (timepoint) |
|
Patel 2013 |
knee OA |
saline |
Partial |
3× |
double |
WOMAC total (6 mo) |
|
Filardo 2015 |
knee OA |
HA |
None |
— |
double |
IKDC (12 mo) |
|
Cole 2017 |
knee OA |
HA |
Partial |
— |
double |
WOMAC pain (12 mo) |
|
Gormeli 2017 |
knee OA |
HA and saline |
None |
— |
double |
IKDC (6 mo) |
|
Di-Martino 2019 |
knee OA |
HA |
None |
— |
double |
IKDC (64 mo) |
|
Bennell 2021 |
knee OA |
saline |
Full |
— |
double |
knee pain NRS change (12 mo) |
|
Chu 2022 |
knee OA |
saline |
Partial |
4.3× |
double |
WOMAC total (60 mo) |
|
Elksnins 2020 |
knee OA |
corticosteroid |
Full |
— |
open |
VAS (12 mo) |
|
Raeissadat 2015 |
knee OA |
HA |
Full |
5.2× |
open |
WOMAC total (12 mo) |
|
Rayegani 2014 |
knee OA |
exercise |
Full |
5.68× |
open |
WOMAC pain (6 mo) |
|
Duymus 2017 |
knee OA |
HA |
None |
— |
unclear |
WOMAC total (12 mo) |
|
Lisi 2018 |
knee OA |
HA |
None |
— |
single |
MRI grade improvement (6 mo) |
|
Louis 2018 |
knee OA |
HA |
Partial |
— |
double |
WOMAC total change (3 mo) |
|
Buendia-Lopez 2018 |
knee OA |
HA |
Full |
— |
unclear |
WOMAC pain (12 mo) |
|
Su 2018 |
knee OA |
HA |
None |
— |
unclear |
WOMAC total (18 mo) |
|
Huang 2019 |
knee OA |
HA |
None |
— |
unclear |
WOMAC total (12 mo) |
|
Lin 2019 |
knee OA |
saline |
Partial |
— |
double |
WOMAC transformed (12 mo) |
|
Yurtbay 2022 |
knee OA |
saline |
Partial |
— |
double |
KOOS (24 mo) |
|
Smith 2016 |
knee OA |
saline |
Partial |
— |
double |
WOMAC total (12 mo) |
|
Cerza 2012 |
knee OA |
HA |
Partial |
— |
open |
WOMAC total (6 mo) |
|
Spakova 2012 |
knee OA |
HA |
Partial |
4.5× |
unclear |
WOMAC total (6 mo) |
|
Paterson 2016 |
knee OA |
HA |
Full |
— |
double |
VAS (3 mo) |
|
Forogh 2016 |
knee OA |
corticosteroid |
None |
— |
double |
KOOS (6 mo) |
|
Montanez-Heredia 2016 |
knee OA |
HA |
Partial |
— |
double |
VAS responders (6 mo) |
|
Filardo 2012 |
knee OA |
HA |
Full |
5× |
double |
IKDC (12 mo) |
|
Di-Martino 2022 |
knee OA |
LP-PRP |
Full |
— |
double |
IKDC (12 mo) |
|
Vaquerizo 2013 |
knee OA |
HA |
None |
— |
unclear |
WOMAC responders (12 mo) |
|
Mishra 2014 |
lateral epicondylitis |
needling anesthetic |
Partial |
— |
double |
VAS percent improvement (6 mo) |
|
Peerbooms 2010 |
lateral epicondylitis |
corticosteroid |
None |
— |
double |
VAS success rate (12 mo) |
|
Gosens 2011 |
lateral epicondylitis |
corticosteroid |
Partial |
— |
double |
VAS DASH success (24 mo) |
|
Krogh 2013 |
lateral epicondylitis |
saline |
None |
— |
double |
PRTEE MD (3 mo) |
|
Behera 2015 |
lateral epicondylitis |
bupivacaine |
Partial |
— |
unclear |
VAS percent improvement (12 mo) |
|
Palacio 2016 |
lateral epicondylitis |
anesthetic and dexamethasone |
Full |
— |
unclear |
PRTEE (6 mo) |
|
Linnanmaki 2020 |
lateral epicondylitis |
saline |
Full |
1.99× |
double |
VAS MD (12 mo) |
|
Randelli 2011 |
rotator cuff repair |
repair alone |
None |
— |
double |
Constant (24 mo) |
|
Rha 2013 |
rotator cuff tendinopathy |
dry needling |
None |
— |
double |
SPADI (6 mo) |
|
Kesikburun 2013 |
rotator cuff tendinopathy |
saline |
None |
— |
double |
WORC (12 mo) |
|
Schwitzguebel 2019 |
supraspinatus interstitial tear |
saline |
None |
— |
double |
lesion volume change mm³ (7 mo) |
|
Pandey 2016 |
rotator cuff repair |
repair alone |
Partial |
— |
unclear |
Constant (24 mo) |
|
Snow 2020 |
rotator cuff repair |
saline |
Partial |
— |
double |
ASES (12 mo) |
|
Cai 2019 |
partial rotator cuff tear |
saline |
Full |
— |
double |
Constant (12 mo) |
|
de Vos 2010 |
achilles tendinopathy |
saline |
None |
— |
double |
VISA-A change (6 mo) |
|
Monto 2014 |
plantar fasciitis |
corticosteroid |
None |
— |
unclear |
AOFAS (24 mo) |
|
Mahindra 2016 |
plantar fasciitis |
corticosteroid |
None |
— |
double |
VAS (3 mo) |
|
Peerbooms 2019 |
plantar fasciitis |
corticosteroid |
None |
— |
double |
FFI pain MD (12 mo) |
|
Acosta-Olivo 2017 |
plantar fasciitis |
corticosteroid |
Partial |
— |
single |
VAS (4 mo) |
|
Kearney 2021 |
achilles tendinopathy |
sham |
Full |
— |
single |
VISA-A (6 mo) |
|
Boesen 2017 |
achilles tendinopathy |
placebo saline |
None |
— |
double |
VISA-A change (6 mo) |
|
Kearney 2013 |
achilles tendinopathy |
eccentric exercise |
Full |
— |
unclear |
VISA-A (6 mo) |
|
Driver 2006 |
diabetic foot ulcer |
saline gel |
None |
— |
single |
healing rate percent (3 mo) |
|
Game 2018 |
diabetic foot ulcer |
standard care |
Partial |
— |
single |
healing rate percent (5 mo) |
|
Li 2015 |
diabetic chronic ulcer |
standard care |
None |
— |
unclear |
healing grade 1 percent (3 mo) |
|
Saad Setta 2011 |
diabetic foot ulcer |
platelet poor plasma |
Full |
— |
unclear |
time to healing (final follow-up) |
|
Kakagia 2007 |
diabetic foot ulcer |
protease modulating matrix |
Partial |
— |
unclear |
ulcer dimension reduction (2 mo) |
|
Gentile 2015 |
androgenetic alopecia |
placebo half head |
Full |
— |
single |
hair density change per cm² (3 mo) |
|
Alves 2016 |
androgenetic alopecia |
placebo half head |
None |
— |
double |
hair density per cm² (6 mo) |
|
Gkini 2014 |
androgenetic alopecia |
none uncontrolled |
Full |
— |
open |
hair density per cm² (12 mo) |
Table 7: Randomised controlled trials of PRP (k = 57). Preparation reporting was graded Full when centrifugation parameters and product characterisation were both given, Partial when one was given, and None when neither was. “Stated fold” is the platelet concentration factor reported by the trial itself.
Where a trial contributed several outcomes, the primary outcome row is shown.
Figure 6: Composition and dose comparisons (A) and preparation reporting completeness (B). In panel A each estimate is scaled to its own range so that mean differences and standardised mean differences can be displayed together; the numeric estimate and metric are printed beside each row. Panel B shows that the majority of randomised trials cannot be linked to a defined product.
Pooled clinical effect where arm-level data permit pooling
Only 5 trials reported complete arm-level means, standard deviations and sample sizes on a common continuous scale. Pooling those under a random-effects model gave a Hedges g of 0.92 (95% CI 0.38–1.46, z = 3.33, p = 0.0009) in favour of PRP, with substantial heterogeneity (τ² = 0.303, Q = 22.37, I² = 82.1%). A sensitivity analysis adding the two trials that reported standard errors in place of standard deviations raised the estimate to 1.46 (95% CI 0.68–2.24) with I² = 93.1% — an illustration of how sensitive this literature is to a single reporting convention, and the reason those two trials were excluded from the primary estimate.
This pooled estimate should not be read as the effect size of PRP. It is the effect size obtainable from the small minority of trials whose reporting permits arithmetic, in a literature where the majority of trials do not report enough to be included. Its main value is as a measure of how little of the published evidence base is actually poolable at the participant level.
Figure 7: Heterogeneity across the published meta-analytic literature (A) and the pooled effect from trials with complete arm-level data (B). Panel A shows the distribution of reported I² across 111 pooled estimates, with the conventional substantial (50%) and considerable (75%) thresholds marked. Panel B is a random-effects forest plot; the diamond is the pooled estimate.
What the existing meta-analytic literature already shows
Across 190 pooled estimates from 36 systematic reviews, 128 of 186 estimates with confidence intervals (68.8%) excluded the null, and 120 favoured PRP against 57 favouring neither arm. For knee osteoarthritis, WOMAC total mean differences had a median of -11.45 points (range -38.69 to -2.09, k = 14) and VAS pain a median of -1.12 points (range -4.95 to -0.18, k = 17). Standardised mean differences across all indications had a median of -0.22 (k = 29), with 62% favouring PRP.
|
Review |
Indication |
Comparison |
Outcome |
Months |
k trials |
Estimate (95% CI) |
I² |
|
Han 2019 |
knee OA |
PRP vs HA |
WOMAC total |
6 |
7 |
-10.78 (-17.51, -4.04) |
94% |
|
Han 2019 |
knee OA |
PRP vs HA |
WOMAC total |
12 |
4 |
-12.11 (-20.21, -4.01) |
94% |
|
McLarnon 2021 |
knee OA |
PRP vs corticosteroid |
WOMAC total |
6 |
4 |
-9.51 (-15.20, -3.83) |
n.a. |
|
Bensa 2025 |
knee OA |
PRP vs placebo |
WOMAC total |
1 |
n.a. |
-8.15 (-13.75, -2.54) |
n.a. |
|
Bensa 2025 |
knee OA |
PRP vs placebo |
WOMAC total |
3 |
n.a. |
-15.90 (-22.98, -8.81) |
n.a. |
|
Bensa 2025 |
knee OA |
PRP vs placebo |
WOMAC total |
6 |
n.a. |
-15.32 (-21.94, -8.69) |
n.a. |
|
Bensa 2025 |
knee OA |
PRP vs placebo |
WOMAC total |
12 |
n.a. |
-14.69 (-25.89, -3.50) |
n.a. |
|
Bensa 2025 |
knee OA |
high-dose PRP vs placebo |
WOMAC total |
12 |
n.a. |
-21.49 (-31.66, -11.32) |
n.a. |
|
Han 2019 |
knee OA |
PRP vs HA |
VAS pain |
12 |
3 |
-4.95 (-7.83, -2.06) |
99% |
|
McLarnon 2021 |
knee OA |
PRP vs corticosteroid |
VAS pain |
6 |
n.a. |
-0.97 (-1.94, 0.00) |
n.a. |
|
Bensa 2025 |
knee OA |
PRP vs placebo |
VAS pain |
3 |
n.a. |
-1.43 (-2.16, -0.71) |
n.a. |
|
Bensa 2025 |
knee OA |
PRP vs placebo |
VAS pain |
6 |
n.a. |
-1.65 (-2.33, -0.97) |
n.a. |
|
Bensa 2025 |
knee OA |
PRP vs placebo |
VAS pain |
12 |
n.a. |
-1.12 (-1.99, -0.25) |
n.a. |
|
Bensa 2025 |
knee OA |
high-dose PRP vs placebo |
VAS pain |
3 |
n.a. |
-1.94 (-2.82, -1.06) |
n.a. |
|
Bensa 2025 |
knee OA |
high-dose PRP vs placebo |
VAS pain |
6 |
n.a. |
-2.10 (-3.22, -0.98) |
n.a. |
|
Bensa 2025 |
knee OA |
high-dose PRP vs placebo |
VAS pain |
12 |
n.a. |
-1.95 (-3.29, -0.60) |
n.a. |
Table 8: Selected pooled estimates for knee osteoarthritis from the umbrella stream. Estimates are mean differences on the WOMAC total and VAS pain scales; negative values favour PRP. Reviews overlap in their constituent trials, so these estimates are shown as a distribution and were not re-pooled.
Figure 8: Pooled mean differences for knee osteoarthritis. WOMAC total (A) and VAS pain (B) from the umbrella stream. Teal markers indicate estimates whose confidence interval excludes the null; grey markers indicate null estimates. The consistency of direction alongside the spread of magnitude is characteristic of this literature.
Figure 9: Pooled standardised mean differences across all indications. All 29 SMD estimates from the umbrella stream, ordered by magnitude. Negative values favour PRP for pain and symptom outcomes; positive values favour PRP where the source review coded improvement positively (wounds, alopecia, and injection-number comparisons).
Composition and dose: what actually moves the outcome
|
Review |
Indication |
Comparison |
Outcome |
Months |
Estimate (95% CI) |
Favours |
|
McLarnon 2021 |
knee OA |
LP-PRP vs steroid |
pain |
6 |
-0.61 (-0.92, -0.29) SMD |
PRP |
|
McLarnon 2021 |
knee OA |
LR-PRP vs steroid |
pain |
6 |
-0.98 (-1.79, -0.17) SMD |
PRP |
|
McLarnon 2021 |
knee OA |
single injection PRP vs steroid |
pain |
6 |
-0.52 (n.a., n.a.) SMD |
neither |
|
McLarnon 2021 |
knee OA |
triple injection PRP vs steroid |
pain |
6 |
-0.94 (-1.31, -0.58) SMD |
PRP |
|
Bensa 2025 |
knee OA |
high-dose PRP vs placebo |
WOMAC total |
12 |
-21.49 (-31.66, -11.32) MD |
PRP |
|
Bensa 2025 |
knee OA |
high-dose PRP vs placebo |
VAS pain |
3 |
-1.94 (-2.82, -1.06) MD |
PRP |
|
Bensa 2025 |
knee OA |
high-dose PRP vs placebo |
VAS pain |
6 |
-2.10 (-3.22, -0.98) MD |
PRP |
|
Bensa 2025 |
knee OA |
high-dose PRP vs placebo |
VAS pain |
12 |
-1.95 (-3.29, -0.60) MD |
PRP |
|
Bensa 2025 |
knee OA |
low-dose PRP vs placebo |
VAS pain |
6 |
-1.28 (-2.44, -0.12) MD |
PRP |
|
Xu 2026 |
knee OA |
LP-PRP vs placebo |
WOMAC function |
6 |
-15.26 (-21.91, -8.61) MD |
PRP |
|
Xu 2026 |
knee OA |
L-PRP vs LP-PRP |
WOMAC function |
6 |
-0.45 (-6.44, 5.53) MD |
neither |
|
Xu 2026 |
knee OA |
LP-PRP vs placebo |
pain |
6 |
-11.07 (-19.56, -2.57) MD |
PRP |
|
Xu 2026 |
knee OA |
L-PRP vs LP-PRP |
pain |
6 |
-0.11 (-1.60, 1.38) MD |
neither |
|
Xu 2026 |
knee OA |
LP-PRP vs placebo |
WOMAC function |
12 |
7.32 (-5.17, 19.80) MD |
neither |
|
Nie 2021 |
knee OA |
LP-PRP vs LR-PRP indirect |
pain |
n.a. |
-0.33 (-1.02, 0.36) SMD |
neither |
|
Nie 2021 |
knee OA |
LP-PRP vs LR-PRP indirect |
function |
n.a. |
0.21 (-0.56, 0.98) SMD |
neither |
|
Peng 2022 |
knee OA |
LR-PRP vs HA |
WOMAC total |
6 |
-4.29 (-7.66, -0.91) MD |
PRP |
|
Peng 2022 |
knee OA |
LR-PRP vs HA |
WOMAC total |
12 |
-6.45 (-19.37, 6.47) MD |
neither |
Table 9: Composition, dose and injection-number comparisons from the umbrella stream. Leukocyte-rich versus leukocyte-poor comparisons are consistently null; platelet dose comparisons consistently favour the higher-dose arm in knee osteoarthritis. Injection-number comparisons favour repeated injection but with wide intervals.
Two patterns separate cleanly. First, the leukocyte question — the most-debated composition variable in the field — is empirically quiet: direct leukocyte-rich versus leukocyte-poor comparisons cross the null in every pooled estimate retrieved, and a published meta-regression of leukocyte content against outcome was likewise null. Second, platelet dose is not quiet. High-concentration PRP outperformed the pooled overall PRP estimate on WOMAC total at 12 months by a wide margin, and a dose-stratified review found a positive gradient across platelet-dose bands in knee osteoarthritis. A parallel analysis in rotator cuff disease found no such gradient, so the dose signal is indication-specific rather than universal.
This is the pivot of the whole analysis. The laboratory stream shows that spin architecture and cumulative dose — not g-force — determine platelet dose, with a between-protocol range of 1.47× to 9.50×. The clinical stream shows that platelet dose is the one composition variable that tracks outcome. And the reporting stream shows that only 7 of 57 trials tell us what dose they gave. The mechanism linking preparation to outcome is not missing; it is unmeasured.
Discussion
Principal findings
Four findings emerge. First, the pooled platelet concentration factor across the published record is 4.63×, not the 3–5× usually quoted, and the spread between protocols (1.47–9.50×) is larger than most clinicians assume. Second, the relative centrifugal force of an individual spin has no measurable relationship with yield across the range in clinical use (R² < 0.01 for both spins), whereas spin architecture has a large one (Hedges g = 1.20). Third, the evidence that clinically used centrifugation destroys platelets is weaker than its rhetorical prominence: the studies that measured integrity directly mostly found none, and several apparent damage findings are recovery losses that are equally consistent with mechanical entrapment. Fourth, the composition variable that tracks clinical outcome is platelet dose, not leukocyte class — and it is the variable least often reported.
Why g-force fails as a predictor
The null g-force result is not a statistical artefact of a noisy literature; it has a mechanical explanation. Sedimentation under centrifugation is governed by force multiplied by time, and by the geometry of the separation — rotor type, tube fill, plasma column height, the position from which the operator aspirates. A protocol at 200 g for 20 minutes and one at 800 g for 5 minutes deliver comparable cumulative dose and, in this dataset, comparable yield. Reporting only the g-force therefore discards most of the mechanically relevant information. The finding that cumulative g·min performed better than either spin force alone, while still explaining only about one seventh of the variance, points to the remaining share being operator technique and device geometry — the parts of the procedure that no protocol description currently captures.
The practical implication is uncomfortable for kit marketing, which usually foregrounds a proprietary speed. Two kits quoting different g-forces may produce indistinguishable products, while two operators using the same kit may not.
The lysis question, reconsidered
The 3,000 g threshold has become a citation habit. It traces to a single viability study, and while that study is methodologically sound, the condition it tested — 3,000 g sustained for 20 minutes — sits at the outer edge of clinical practice. Against it stand four independent direct measurements of platelet integrity at routine forces that found nothing: P-selectin unchanged during concentration, no CD62P difference between rotor geometries, lactate dehydrogenase unchanged at 1,600 g for 20 minutes, and no difference in activated-platelet fraction between protocols. A historical haematology protocol using 3,000 g for 20 minutes reported 86% platelet harvest without viability loss.
This does not mean centrifugation is biologically inert. Activation markers do rise at 800–1,200 g in at least one careful series, aggregation and morphological change have been documented at 1,000 g on the second spin, and electron microscopy of one double-spin system showed aggregate alteration. The honest summary is that partial, sub-lytic activation is real and force-related, while frank lysis at clinically used settings is not established. Since partial activation may release growth factors prematurely into the supernatant rather than at the treatment site, this remains clinically relevant — but it is a different mechanism from destruction, and it demands different assays.
Growth factors are the wrong reporting target
Growth factor concentrations varied by two to three orders of magnitude between studies and did not correlate significantly with platelet concentration factor. Some of that variance is biological, but much is methodological: whether the sample was activated before assay, whether platelets were lysed to release intracellular stores, which immunoassay platform was used, and whether results were normalised per volume or per million platelets. Until those choices are standardised, reported growth factor concentrations are not comparable across laboratories and should not be used to characterise a product. Platelet dose — the absolute number of platelets delivered, in billions — is measurable with a standard haematology analyser, reproducible, and directly linked in the dose-stratified clinical literature to outcome.
The reporting failure is the rate-limiting step
The most actionable finding in this analysis is not biological. 23 of 57 randomised trials (40%) provide no usable preparation detail, and only 7 state the platelet concentration factor delivered. Meta-analyses of this literature routinely report I² above 75% — in this umbrella stream, 45% of pooled estimates exceeded that threshold — and then attribute the heterogeneity to preparation differences they cannot observe. That attribution may well be correct. It is simply untestable with the data as published, and it will remain untestable however many further trials are run under current reporting norms.
Several classification frameworks already exist for this purpose, reporting platelet dose, purity, activation state and efficiency. Their existence is not the problem; their adoption is. A minimum dataset of six numbers — whole-blood platelet count, final platelet concentration, final volume, absolute platelet dose in billions, leukocyte concentration, and spin parameters including rotor type — would take a trialist minute to report and would transform the interpretability of the entire field.
Limitations
- The laboratory stream pools preparation arms from heterogeneous protocols and donor populations rather than randomised comparisons; between-arm contrasts are therefore observational and susceptible to confounding by laboratory, device and donor characteristics.
- Several studies contribute multiple arms, which introduces within-study correlation not modelled in the descriptive pooling; the between-architecture contrast was confirmed with a rank-based test to reduce sensitivity to this.
- Only 5 randomised trials reported sufficient arm-level data for participant-level pooling, so the pooled clinical effect is imprecise and not representative of the full trial literature.
- Umbrella estimates overlap substantially in their constituent trials and were deliberately not re-pooled; the distributional summary presented instead cannot be interpreted as a single effect size.
- Growth factor comparisons are limited by assay heterogeneity, and the small number of arms reporting each analyte limits the power of the correlation analyses.
- Publication and language bias were not formally assessed for the laboratory stream, where funnel-plot methods are not applicable to single-arm descriptive data.
Implications for practice and research
For clinicians, three recommendations follow from these data. Choose a double-spin or buffy-coat protocol when platelet dose matters, because architecture, not speed, drives yield. Stop selecting kits on quoted g-force. Measure and record the platelet count of the product delivered — a single analyser run — so that individual practice can be audited against the dose-response literature.
For researchers, the priority is not another placebo-controlled trial of unspecified PRP. It is dose-stratified randomisation: the same preparation system, deliberately delivering defined platelet doses, in a single indication, with product characterization reported for every participant. The dose signal already visible in the pooled literature is strong enough to justify that design and too confounded to be settled without it.
Conclusion
Centrifugation determines what is in the syringe. Spin architecture and cumulative centrifugal dose govern platelet concentration — pooled at 4.63× across the published record, ranging from 1.47× to 9.50× — while the raw g-force of any single spin does not. The widely repeated claim that clinically used centrifugation lyses platelets is not supported by direct integrity assays, although sub-lytic activation at higher forces is real. Growth factor concentrations are too assay-dependent to serve as a product descriptor. In the clinical literature, PRP outperforms its comparators more often than not, and platelet dose is the only composition variable with a consistent signal — yet 23 of 57 randomised trials do not report their preparation at all. The chain from centrifuge to patient is not broken at the biology. It is broken at the methods section, and it can be repaired with six numbers.
Declarations
Not applicable. This study analysed previously published data only.
Availability of data and materials
All extracted data rows, with their source bindings, are available from the corresponding author on reasonable request.
Competing interests
The authors declare that they have no competing interests.
Funding
No funding.
Authors' contributions
[MHK, CAPMR] conceived and designed the study. All authors read, revised and approved the final manuscript.
References
- Perez AG, Lichy R, Lana JF, Rodrigues AA, Luzo AC, et al. (2013) Prediction and modulation of platelet recovery by discontinuous centrifugation of whole blood for the preparation of pure platelet-rich plasma. Biores Open Access. 2(4):307-14.
- Perez AG, Lana JF, Rodrigues AA, Luzo AC, Belangero WD, et al. (2014) Relevant aspects of centrifugation step in the preparation of platelet-rich plasma. ISRN Hematol. 2014:176060.
- Amable PR, Carias RB, Teixeira MV, da Cruz Pacheco I, Correa do Amaral RJ, et al. (2013) Platelet-rich plasma preparation for regenerative medicine: optimization and quantification of cytokines and growth factors. Stem Cell Res Ther. 4(3):67.
- Bausset O, Giraudo L, Veran J, Magalon J, Coudreuse JM, et al. (2012) Formulation and storage of platelet-rich plasma homemade product. Biores Open Access. 1(3):115-123.
- Fitzpatrick J, Bulsara MK, McCrory PR, Richardson MD, Zheng MH. (2017) Analysis of Platelet-Rich Plasma Extraction: Variations in Platelet and Blood Components Between 4 Common Commercial Kits. Orthop J Sports Med. X5(1):2325967116675272.
- Magalon J, Chateau AL, Bertrand B, Louis ML, Silvestre A, et al.(2016) DEPA classification: a proposal for standardising PRP use and a retrospective application of available devices. BMJ Open Sport Exerc Med. 2(1):e000060.
- Piao L, Piao L, Park H, Jo CH.(2017) Theoretical prediction and validation of cell recovery rates in preparing platelet-rich plasma through a centrifugation. PLoS One. X 12(11):e0187509. Available from:
- Yin W, Qi X, Zhang Y, Sheng J, Xu ZX, et al.(2016) Advantages of pure platelet-rich plasma compared with leukocyte- and platelet-rich plasma in promoting repair of bone defects. J Transl MedX. 14:73.
- Roh YH, Kim W, Park KU, Oh JH.(2016) Cytokine-release kinetics of platelet-rich plasma according to various activation protocols. Bone Joint Res. 5(2):37-45.
- Etulain J, Mena HA, Meiss RP, Frechtel G, Gutt S, et al.(2018) An optimised protocol for platelet-rich plasma preparation to improve its angiogenic and regenerative properties. Sci Rep 8(1):1513.
- Taniguchi Y, Yoshioka T, Sugaya H, Gosho M, Aoto K, et al.(2019) Growth factor levels in leukocyte-poor platelet-rich plasma and correlations with donor age, gender, and platelets in the Japanese population. J Exp Orthop. 6(1):4.
- Dejnek M, Witkowski J, Moreira H, Płaczkowska S, Morasiewicz P, et al.(2022)Content of blood cell components, inflammatory cytokines and growth factors in autologous platelet-rich plasma obtained by various methods. World J Orthop. 13(6):587-602.
- Muthu S, Krishnan A, Ramanathan KR.(2022) Standardization and validation of a conventional high yield platelet-rich plasma preparation protocol. Ann Med Surg (Lond). 82:104593.
- Machado ES, Soares FP, Yamaguchi RS, Felipone WK, Meves R, et al.(2022) A Simple Double-Spin Closed Method for Preparing Platelet-Rich Plasma. Cureus.;14(1):e20899.
- Carvalho A, Ferreira AF, Soares M, Santos S, Tomé P, et al.(2024) Optimization of Platelet-Rich Plasma Preparation for Regenerative Medicine: Comparison of Different Anticoagulants and Resuspension Media. Bioengineering (Basel). 11(3):209.
- Husak V, Povelychenko O, Maltseva V, Romanenko K, Vorontsov P, et al.(2025)Growth factor and cell content in Platelet-Rich Plasma (PRP), Leukocyte- and Platelet-Rich Plasma (L-PRP), Platelet-Rich Fibrin (PRF) in patients with long bone defects from combat injuries. BMC Musculoskelet Disord. 26(1):1106.
- Miroshnychenko O, Chalkley RJ, Leib RD, Everts PA, Dragoo JL.(2020) Proteomic analysis of platelet-rich and platelet-poor plasma. Regen Ther. 15:226-235.
- Huber SC, Cunha Júnior JL, Montalvão S, da Silva LQ, Paffaro AU, et al.(2016) In vitro study of the role of thrombin in platelet rich plasma (PRP) preparation: utility for gel formation and impact in growth factors release. J Stem Cells Regen Med. 12(1):2-9.
- Kruel AVS, Ferreira M, Agostini D, Diesel CV, Queiroz M, et al.(2026) Comparative Characterization of Leukocyte-Rich Platelet-Rich Plasma (L-PRP) and Injectable Platelet-Rich Fibrin (i-PRF): A Laboratory Study. Cells. 15(10):886.
- Garrison L, Nicks D, Nicks S, Green J, Pham H, et al.(2026)Evaluation of accuracy of platelet rich plasma preparation devices for cardiac surgery: A prospective, randomized, single blinded study. J Extra Corpor Technol. 58(2):124-127.
- Panero AJ, Warrick AE, Nakagawa H, Andersen W, Hirahara A, et al.(2025)Comparing Biologic Factor Composition of Birth Tissue-Derived Allografts and Platelet-Rich Plasma. Cureus. 17(8):e90259.
- Okumo T, Sato A, Izukashi K, Ohta M, Oike J, et al.(2023) Multifactorial Comparative Analysis of Platelet-Rich Plasma and Serum Prepared Using a Commercially Available Centrifugation Kit. Cureus. 15(11):e48918.
- Menon SS, Balakrishnan B, Kurumathur Vasudevan A, Rajan Peter M, Suresh R.(2026) Effect of Reduced Centrifugation Time on Leukocyte, Platelet, and Growth Factor Levels in Platelet-Rich Fibrin (PRF) Prepared Using Low-Speed Relative Centrifugal Force (RCF): An Ex Vivo Study. Cureus. 2026;18(1):e100846.
- Araki J, Jona M, Eto H, Aoi N, Kato H, et al.(2012) Optimized preparation method of platelet-concentrated plasma and noncoagulating platelet-derived factor concentrates: maximization of platelet concentration and removal of fibrinogen. Tissue Eng Part C Methods. 18(3):176-185.
- Mazzocca AD, McCarthy MB, Chowaniec DM, Cote MP, Romeo AA, et al.(2012) Platelet-rich plasma differs according to preparation method and human variability. J Bone Joint Surg Am. 94(4):308-316.
- Weibrich G, Kleis WK, Hafner G, Hitzler WE.(2002) Growth factor levels in platelet-rich plasma and correlations with donor age, sex, and platelet count. J Craniomaxillofac Surg. 30(2):97-102.
- Eppley BL, Woodell JE, Higgins J.(2004) Platelet quantification and growth factor analysis from platelet-rich plasma: implications for wound healing. Plast Reconstr Surg. X 114(6):1502-1508.
- Sundman EA, Cole BJ, Fortier LA.(2011) Growth factor and catabolic cytokine concentrations are influenced by the cellular composition of platelet-rich plasma. Am J Sports Med. X 39(10):2135-2140.
- Jo CH, Roh YH, Kim JE, Shin S, Yoon KS.(2013) Optimizing platelet-rich plasma gel formation by varying time and gravitational forces during centrifugation. J Oral Implantol. 39(5):525-532.
- Slichter SJ, Harker LA.(1976) Preparation and storage of platelet concentrates. I. Factors influencing the harvest of viable platelets from whole blood. Br J Haematol. 34(3):395-402.
- Tamimi FM, Montalvo S, Tresguerres I, Blanco Jerez L.(2007) A comparative study of 2 methods for obtaining platelet-rich plasma. J Oral Maxillofac Surg. 65(6):1084-1093.
- Oh JH, Kim W, Park KU, Roh YH.(2015) Comparison of the Cellular Composition and Cytokine-Release Kinetics of Various Platelet-Rich Plasma Preparations. Am J Sports Med. 43(12):3062-3070.
- Kobayashi Y, Saita Y, Nishio H, Ikeda H, Takazawa Y, et al.(2016) Leukocyte concentration and composition in platelet-rich plasma (PRP) influences the growth factor and protease concentrations. J Orthop Sci. X;21(5):683-689.
- Magalon J, Bausset O, Serratrice N, Giraudo L, Aboudou H, et al.(2014) Characterization and comparison of 5 platelet-rich plasma preparations in a single-donor model. Arthroscopy. 30(5):629-638.
- Weibrich G, Kleis WK, Streckbein P, Moergel M, Hitzler WE, et al.(2012) Comparison of point-of-care methods for preparation of platelet concentrate (platelet-rich plasma). Int J Oral Maxillofac Implants. 27(4):762-769.
- Ozer K, Kankaya Y, Colak O, Kocer U.(2019) The Impact of Duration and Force of Centrifugation on Platelet Content and Mass in the Preparation of Platelet-Rich Plasma. Aesthetic Plast Surg. 43(4):1078-1084.
- Xie M, Zhao B, Teng J, Yang L, Wang Z, Xu S, et al. Effect of Fixed-Angle and Horizontal Rotor Centrifugation on Optimized Platelet-Rich Plasma Preparation. Aesthetic Plast Surg. 2025;49(17):4989-4998.
- Croisé B, Paré A, Joly A, Louisy A, Laure B, et al.(2020) Optimized centrifugation preparation of the platelet rich plasma: Literature review. J Stomatol Oral Maxillofac Surg.121(2):150-154.
- Dohan Ehrenfest DM, Rasmusson L, Albrektsson T. (2009) Classification of platelet concentrates: from pure platelet-rich plasma (P-PRP) to leucocyte- and platelet-rich fibrin (L-PRF). Trends Biotechnol.27(3):158-167.
- Landesberg R, Roy M, Glickman RS.(2005) Quantification of growth factor levels using a simplified method of platelet-rich plasma gel preparation. J Oral Maxillofac Surg. 58(3):297-300; discussion 300-1.
- Patel S, Dhillon MS, Aggarwal S, Marwaha N, Jain A. (2013) Treatment with platelet-rich plasma is more effective than placebo for knee osteoarthritis: a prospective, double-blind, randomized trial. Am J Sports Med. 41(2):356-364.
- Filardo G, Di Matteo B, Di Martino A, Merli ML, Cenacchi A, et al. (2015) Platelet-Rich Plasma Intra-articular Knee Injections Show No Superiority Versus Viscosupplementation: A Randomized Controlled Trial. Am J Sports Med. 43(7):1575-1582.
- Cole BJ, Karas V, Hussey K, Pilz K, Fortier LA. (2017) Hyaluronic Acid Versus Platelet-Rich Plasma: A Prospective, Double-Blind Randomized Controlled Trial Comparing Clinical Outcomes and Effects on Intra-articular Biology for the Treatment of Knee Osteoarthritis. Am J Sports Med. 45(2):339-346.
- Görmeli G, Görmeli CA, Ataoglu B, Çolak C, Aslantürk O, et al. (2017)Multiple PRP injections are more effective than single injections and hyaluronic acid in knees with early osteoarthritis: a randomized, double-blind, placebo-controlled trial. Knee Surg Sports Traumatol Arthrosc. 25(3):958-965.
- Di Martino A, Di Matteo B, Papio T, Tentoni F, Selleri F, et al. Platelet-Rich Plasma Versus Hyaluronic Acid Injections for the Treatment of Knee Osteoarthritis: Results at 5 Years of a Double-Blind, Randomized Controlled Trial. Am J Sports Med. 2019;47(2):347-354.
- Bennell KL, Paterson KL, Metcalf BR, Duong V, Eyles J, et al. (2021)Effect of Intra-articular Platelet-Rich Plasma vs Placebo Injection on Pain and Medial Tibial Cartilage Volume in Patients With Knee Osteoarthritis: The RESTORE Randomized Clinical Trial. JAMA. 326(20):2021-2030. Available from: https://doi.org/10.1001/jama.2021.19415
- Chu J, Duan W, Yu Z, Tao T, Xu J, et al. (2022) Intra-articular injections of platelet-rich plasma decrease pain and improve functional outcomes than sham saline in patients with knee osteoarthritis. Knee Surg Sports Traumatol Arthrosc. 30(12):4063-4071.
- Elksniņš-Finogejevs A, Vidal L, Peredistijs A.(2020) Intra-articular platelet-rich plasma vs corticosteroids in the treatment of moderate knee osteoarthritis: a single-center prospective randomized controlled study with a 1-year follow up. J Orthop Surg Res. 15(1):257.
- Raeissadat SA, Rayegani SM, Hassanabadi H, Fathi M, Ghorbani E, et al. (2015) Knee Osteoarthritis Injection Choices: Platelet- Rich Plasma (PRP) Versus Hyaluronic Acid (A one-year randomized clinical trial). Clin Med Insights Arthritis Musculoskelet Disord. 8:1-8.
- Rayegani SM, Raeissadat SA, Taheri MS, Babaee M, Bahrami MH, et al. (2014) Does intra articular platelet rich plasma injection improve function, pain and quality of life in patients with osteoarthritis of the knee? A randomized clinical trial. Orthop Rev (Pavia). 6(3):5405.
- Duymus TM, Mutlu S, Dernek B, Komur B, Aydogmus S, et al. (2017)Choice of intra-articular injection in treatment of knee osteoarthritis: platelet-rich plasma, hyaluronic acid or ozone options. Knee Surg Sports Traumatol Arthrosc. 25(2):485-492.
- Lisi C, Perotti C, Scudeller L, Sammarchi L, Dametti F, et al.(2018) Treatment of knee osteoarthritis: platelet-derived growth factors vs. hyaluronic acid. A randomized controlled trial. Clin Rehabil. 32(3):330-339.
- Louis ML, Magalon J, Jouve E, Bornet CE, Mattei JC, Chagnaud C, et al. Growth Factors Levels Determine Efficacy of Platelets Rich Plasma Injection in Knee Osteoarthritis: A Randomized Double Blind Noninferiority Trial Compared With Viscosupplementation. Arthroscopy. 34(5):1530-1540.e2.
- Buendía-López D, Medina-Quirós M, Fernández-Villacañas Marín MÁ. (2018) Clinical and radiographic comparison of a single LP-PRP injection, a single hyaluronic acid injection and daily NSAID administration with a 52-week follow-up: a randomized controlled trial. J Orthop Traumatol. 19(1):3.
- Su K, Bai Y, Wang J, Zhang H, Liu H, et al. (2018)Comparison of hyaluronic acid and PRP intra-articular injection with combined intra-articular and intraosseous PRP injections to treat patients with knee osteoarthritis. Clin Rheumatol. 37(5):1341-1350.
- Huang Y, Liu X, Xu X, Liu J. (2019) Intra-articular injections of platelet-rich plasma, hyaluronic acid or corticosteroids for knee osteoarthritis: A prospective randomized controlled study. Orthopade. 48(3):239-247.
- Lin KY, Yang CC, Hsu CJ, Yeh ML, Renn JH. (2019) Intra-articular Injection of Platelet-Rich Plasma Is Superior to Hyaluronic Acid or Saline Solution in the Treatment of Mild to Moderate Knee Osteoarthritis: A Randomized, Double-Blind, Triple-Parallel, Placebo-Controlled Clinical Trial. Arthroscopy. 35(1):106-117.
- Yurtbay A, Say F, Çinka H, Ersoy A.(2022) Multiple platelet-rich plasma injections are superior to single PRP injections or saline in osteoarthritis of the knee: the 2-year results of a randomized, double-blind, placebo-controlled clinical trial. Arch Orthop Trauma Surg. 142(10):2755-2768.
- Smith PA. Intra-articular Autologous Conditioned Plasma Injections Provide Safe and Efficacious Treatment for Knee Osteoarthritis: An FDA-Sanctioned, Randomized, Double-blind, Placebo-controlled Clinical Trial. Am J Sports Med.44(4):884-891.
- Cerza F, Carnì S, Carcangiu A, Di Vavo I, Schiavilla V, et al. (2012) Comparison between hyaluronic acid and platelet-rich plasma, intra-articular infiltration in the treatment of gonarthrosis. Am J Sports Med. 40(12):2822-2827.
- Spaková T, Rosocha J, Lacko M, Harvanová D, Gharaibeh A. (2012) Treatment of knee joint osteoarthritis with autologous platelet-rich plasma in comparison with hyaluronic acid. Am J Phys Med Rehabil. 91(5):411-417.
- Paterson KL, Nicholls M, Bennell KL, Bates D. (2016) Intra-articular injection of photo-activated platelet-rich plasma in patients with knee osteoarthritis: a double-blind, randomized controlled pilot study. BMC Musculoskelet Disord. 17:67.
- Forogh B, Mianehsaz E, Shoaee S, Ahadi T, Raissi GR, et al. (2016)Effect of single injection of platelet-rich plasma in comparison with corticosteroid on knee osteoarthritis: a double-blind randomized clinical trial. J Sports Med Phys Fitness. 56(7-8):901-908.
- Montañez-Heredia E, Irízar S, Huertas PJ, Otero E, Del Valle M, et al. (2016) Intra-Articular Injections of Platelet-Rich Plasma versus Hyaluronic Acid in the Treatment of Osteoarthritic Knee Pain: A Randomized Clinical Trial in the Context of the Spanish National Health Care System. Int J Mol Sci. 17(7):E1064.
- Filardo G, Kon E, Di Martino A, Di Matteo B, Merli ML, et al. (2012) Platelet-rich plasma vs hyaluronic acid to treat knee degenerative pathology: study design and preliminary results of a randomized controlled trial. BMC Musculoskelet Disord. 13:229.
- Di Martino A, Boffa A, Andriolo L, Romandini I, Altamura SA, et al. (2012) Leukocyte-Rich versus Leukocyte-Poor Platelet-Rich Plasma for the Treatment of Knee Osteoarthritis: A Double-Blind Randomized Trial. Am J Sports Med. 50(3):609-617.
- Vaquerizo V, Plasencia MÁ, Arribas I, Seijas R, Padilla S, et al. (2013) Comparison of intra-articular injections of plasma rich in growth factors (PRGF-Endoret) versus Durolane hyaluronic acid in the treatment of patients with symptomatic osteoarthritis: a randomized controlled trial. Arthroscopy. 29(10):1635-1643.
- Mishra AK, Skrepnik NV, Edwards SG, Jones GL, Sampson S, et al. (2014) Efficacy of platelet-rich plasma for chronic tennis elbow: a double-blind, prospective, multicenter, randomized controlled trial of 230 patients. Am J Sports Med. 42(2):463-471.
- Peerbooms JC, Sluimer J, Bruijn DJ, Gosens T. (2010) Positive effect of an autologous platelet concentrate in lateral epicondylitis in a double-blind randomized controlled trial: platelet-rich plasma versus corticosteroid injection with a 1-year follow-up. Am J Sports Med. 38(2):255-262.
- Gosens T, Peerbooms JC, van Laar W, den Oudsten BL. (2011) Ongoing positive effect of platelet-rich plasma versus corticosteroid injection in lateral epicondylitis: a double-blind randomized controlled trial with 2-year follow-up. Am J Sports Med. 39(6):1200-1208.
- Krogh TP, Fredberg U, Stengaard-Pedersen K, Christensen R, Jensen P, et al. (2013) Treatment of lateral epicondylitis with platelet-rich plasma, glucocorticoid, or saline: a randomized, double-blind, placebo-controlled trial. Am J Sports Med. 41(3):625-635.
- Behera P, Dhillon M, Aggarwal S, Marwaha N, Prakash M. (2016) Leukocyte-poor platelet-rich plasma versus bupivacaine for recalcitrant lateral epicondylar tendinopathy. J Orthop Surg (Hong Kong). 2015;23(1):6-10.
- Palacio EP, Schiavetti RR, Kanematsu M, Ikeda TM, Mizobuchi RR, Galbiatti JA. Effects of platelet-rich plasma on lateral epicondylitis of the elbow: prospective randomized controlled trial. Rev Bras Ortop. 51(1):90-95.
- Linnanmäki L, Kanto K, Karjalainen T, Leppänen OV, Lehtinen J. (2020) Platelet-rich Plasma or Autologous Blood Do Not Reduce Pain or Improve Function in Patients with Lateral Epicondylitis: A Randomized Controlled Trial. Clin Orthop Relat Res.478(8):1892-1900.
- Randelli P, Arrigoni P, Ragone V, Aliprandi A, Cabitza P. (2011) Platelet rich plasma in arthroscopic rotator cuff repair: a prospective RCT study, 2-year follow-up. J Shoulder Elbow Surg. 20(4):518-528.
- Rha DW, Park GY, Kim YK, Kim MT, Lee SC. (2013) Comparison of the therapeutic effects of ultrasound-guided platelet-rich plasma injection and dry needling in rotator cuff disease: a randomized controlled trial. Clin Rehabil. 27(2):113-122.
- Kesikburun S, Tan AK, Yilmaz B, Yaşar E, Yazicioğlu K. (2013) Platelet-rich plasma injections in the treatment of chronic rotator cuff tendinopathy: a randomized controlled trial with 1-year follow-up. Am J Sports Med. 41(11):2609-2616.
- Schwitzguebel AJ, Kolo FC, Tirefort J, Kourhani A, Nowak A, et al. (2019) Efficacy of Platelet-Rich Plasma for the Treatment of Interstitial Supraspinatus Tears: A Double-Blinded, Randomized Controlled Trial. Am J Sports Med. 47(8):1885-1892.
- Pandey V, Bandi A, Madi S, Agarwal L, Acharya KK, et al. (2016) Does application of moderately concentrated platelet-rich plasma improve clinical and structural outcome after arthroscopic repair of medium-sized to large rotator cuff tear? A randomized controlled trial. J Shoulder Elbow Surg. 25(8):1312-1322.
- Snow M, Hussain F, Pagkalos J, Kowalski T, Green M, et al. (2020) The Effect of Delayed Injection of Leukocyte-Rich Platelet-Rich Plasma Following Rotator Cuff Repair on Patient Function: A Randomized Double-Blind Controlled Trial. Arthroscopy. 36(3):648-657.
- Cai YU, Sun Z, Liao B, Song Z, Xiao T, et al. (2019)Sodium Hyaluronate and Platelet-Rich Plasma for Partial-Thickness Rotator Cuff Tears. Med Sci Sports Exerc. 51(2):227-233.
- de Vos RJ, Weir A, van Schie HT, Bierma-Zeinstra SM, Verhaar JA, et al. (2010) Platelet-rich plasma injection for chronic Achilles tendinopathy: a randomized controlled trial. JAMA. 303(2):144-149.
- Monto RR. (2014) Platelet-rich plasma efficacy versus corticosteroid injection treatment for chronic severe plantar fasciitis. Foot Ankle Int. 35(4):313-318.
- Mahindra P, Yamin M, Selhi HS, Singla S, Soni A. (2016) Chronic Plantar Fasciitis: Effect of Platelet-Rich Plasma, Corticosteroid, and Placebo. Orthopedics. 39(2):e285-9.
- Peerbooms JC, Lodder P, den Oudsten BL, Doorgeest K, Schuller HM, et al.(2019) Positive Effect of Platelet-Rich Plasma on Pain in Plantar Fasciitis: A Double-Blind Multicenter Randomized Controlled Trial. Am J Sports Med. 47(13):3238-3246.
- Acosta-Olivo C, Elizondo-Rodriguez J, Lopez-Cavazos R, Vilchez-Cavazos F, Simental-Mendia M, et al. (2017) Plantar Fasciitis-A Comparison of Treatment with Intralesional Steroids versus Platelet-Rich Plasma A Randomized, Blinded Study. J Am Podiatr Med Assoc. 107(6):490-496.
- Kearney RS, Ji C, Warwick J, Parsons N, Brown J, et al. (2021) Effect of Platelet-Rich Plasma Injection vs Sham Injection on Tendon Dysfunction in Patients With Chronic Midportion Achilles Tendinopathy: A Randomized Clinical Trial. JAMA. 326(2):137-144.
- Boesen AP, Hansen R, Boesen MI, Malliaras P, Langberg H. (2017) Effect of High-Volume Injection, Platelet-Rich Plasma, and Sham Treatment in Chronic Midportion Achilles Tendinopathy: A Randomized Double-Blinded Prospective Study. Am J Sports Med. 45(9):2034-2043.
- Kearney RS, Parsons N, Costa ML. (2013) Achilles tendinopathy management: A pilot randomised controlled trial comparing platelet-richplasma injection with an eccentric loading programme. Bone Joint Res. 2(10):227-232.
- Driver VR, Hanft J, Fylling CP, Beriou JM. (2006) Autologel Diabetic Foot Ulcer Study Group. A prospective, randomized, controlled trial of autologous platelet-rich plasma gel for the treatment of diabetic foot ulcers. Ostomy Wound Manage. x52(6):68-70, 72, 74 passim.
- Game F, Jeffcoate W, Tarnow L, Jacobsen JL, Whitham DJ, et al. (2018) LeucoPatch system for the management of hard-to-heal diabetic foot ulcers in the UK, Denmark, and Sweden: an observer-masked, randomised controlled trial. Lancet Diabetes Endocrinol. 6(11):870-878.
- Li L, Chen D, Wang C, Yuan N, Wang Y, et al. (2015) Autologous platelet-rich gel for treatment of diabetic chronic refractory cutaneous ulcers: A prospective, randomized clinical trial. Wound Repair Regen. X23(4):495-505.
- Saad Setta H, Elshahat A, Elsherbiny K, Massoud K, Safe I. (2011) Platelet-rich plasma versus platelet-poor plasma in the management of chronic diabetic foot ulcers: a comparative study. Int Wound J. 8(3):307-312.
- Kakagia DD, Kazakos KJ, Xarchas KC, Karanikas M, Georgiadis GS, et al. (2007) Synergistic action of protease-modulating matrix and autologous growth factors in healing of diabetic foot ulcers. A prospective randomized trial. J Diabetes Complications. 21(6):387-391.
- Gentile P, Garcovich S, Bielli A, Scioli MG, Orlandi A, et al. (2015)The Effect of Platelet-Rich Plasma in Hair Regrowth: A Randomized Placebo-Controlled Trial. Stem Cells Transl Med. 4(11):1317-1323.
- Alves R, Grimalt R. Randomized Placebo-Controlled, Double-Blind. (2006) Half-Head Study to Assess the Efficacy of Platelet-Rich Plasma on the Treatment of Androgenetic Alopecia. Dermatol Surg. 42(4):491-497.
- Gkini MA, Kouskoukis AE, Tripsianis G, Rigopoulos D, Kouskoukis K. (2014) Study of platelet-rich plasma injections in the treatment of androgenetic alopecia through an one-year period. J Cutan Aesthet Surg. 7(4):213-219.
- Han Y, Huang H, Pan J, Lin J, Zeng L, et al. (2019) Meta-analysis Comparing Platelet-Rich Plasma vs Hyaluronic Acid Injection in Patients with Knee Osteoarthritis. Pain Med. 20(7):1418-1429.
- McLarnon M, Heron N. (2021) Intra-articular platelet-rich plasma injections versus intra-articular corticosteroid injections for symptomatic management of knee osteoarthritis: systematic review and meta-analysis. BMC Musculoskelet Disord. 2021;22(1):550.
- Bensa A, Previtali D, Sangiorgio A, Boffa A, Salerno M, (2025)PRP Injections for the Treatment of Knee Osteoarthritis: The Improvement Is Clinically Significant and Influenced by Platelet Concentration: A Meta-analysis of Randomized Controlled Trials. Am J Sports Med. 53(3):745-754.
- Xu B, Huang X, Su X, Fu Y, Feng S, et al. (2026) Leukocyte-rich versus leukocyte-poor platelet-rich plasma and hyaluronic acid for knee osteoarthritis: a systematic review and network meta-analysis. J Orthop Surg Res. 21(1):222.
- Berrigan W, Tao F, Kopcow J, Park AL, Allen I, et al.(2024) The Effect of Platelet Dose on Outcomes after Platelet Rich Plasma Injections for Musculoskeletal Conditions: A Systematic Review and Meta-Analysis. Curr Rev Musculoskelet Med. 17(12):570-588.
- Nie LY, Zhao K, Ruan J, Xue J. (2021) Effectiveness of Platelet-Rich Plasma in the Treatment of Knee Osteoarthritis: A Meta-analysis of Randomized Controlled Clinical Trials. Orthop J Sports Med. 9(3):2325967120973284.
- Peng YN, Chen JL, Hsu CC, Chen CPC, Suputtitada A. (2022) Intra-Articular Leukocyte-Rich Platelet-Rich Plasma versus Intra-Articular Hyaluronic Acid in the Treatment of Knee Osteoarthritis: A Meta-Analysis of 14 Randomized Controlled Trials. Pharmaceuticals (Basel). 15(8):974.
- Vilchez-Cavazos F, Millán-Alanís JM, Blázquez-Saldaña J, Álvarez-Villalobos N, Peña-Martínez VM, et al. (2019) Comparison of the Clinical Effectiveness of Single Versus Multiple Injections of Platelet-Rich Plasma in the Treatment of Knee Osteoarthritis: A Systematic Review and Meta-analysis. Orthop J Sports Med. 7(12):2325967119887116.
- Zhang Q, Liu T, Gu Y, Gao Y, Ni J. (2022) Efficacy and safety of platelet-rich plasma combined with hyaluronic acid versus platelet-rich plasma alone for knee osteoarthritis: a systematic review and meta-analysis. J Orthop Surg Res. 17(1):499.
- Dai WL, Zhou AG, Zhang H, Zhang J. (2017) Efficacy of Platelet-Rich Plasma in the Treatment of Knee Osteoarthritis: A Meta-analysis of Randomized Controlled Trials. Arthroscopy. 33(3):659-670.e1.
- Xu Z, Luo J, Huang X, Wang B, Zhang J, et al. (2017) Efficacy of Platelet-Rich Plasma in Pain and Self-Report Function in Knee Osteoarthritis: A Best-Evidence Synthesis. Am J Phys Med Rehabil. 96(11):793-800.
- Zhao D, Pan JK, Yang WY, Han YH, Zeng LF, et al. (2021) Intra-Articular Injections of Platelet-Rich Plasma, Adipose Mesenchymal Stem Cells, and Bone Marrow Mesenchymal Stem Cells Associated With Better Outcomes Than Hyaluronic Acid and Saline in Knee Osteoarthritis: A Systematic Review and Network Meta-analysis. Arthroscopy. 2021;
- -Carreño T, Lozano-Paniagua D, Nievas-Soriano BJ, Murillo-Cancho AF, Gázquez-Aguilera EM, et al. (2026) Efficacy and Safety of Platelet-Rich Plasma in Knee Osteoarthritis: Umbrella Meta-Analysis Based on Clinical Evidence, Methodological Quality and Therapeutic Positioning. Clin Pract. 16(4):75.
- Arirachakaran A, Sukthuayat A, Sisayanarane T, Laoratanavoraphong S, Kanchanatawan W, et al. (2016) Platelet-rich plasma versus autologous blood versus steroid injection in lateral epicondylitis: systematic review and network meta-analysis. J Orthop Traumatol. 17(2):101-112.
- Li A, Wang H, Yu Z, Zhang G, Feng S, et al. (2019) Platelet-rich plasma vs corticosteroids for elbow epicondylitis: A systematic review and meta-analysis. Medicine (Baltimore). 98(51):e18358.
- Maroun R, Daher M, Boufadel P, Lopez R, Khan AZ, et al. (2025) Platelet rich plasma versus corticosteroids for lateral epicondylitis: a meta-analysis of randomized clinical trials. Clin Shoulder Elb. 28(1):40-48.
- Simental-Mendía M, Vilchez-Cavazos F, Álvarez-Villalobos N, Blázquez-Saldaña J, Peña-Martínez V, et al. (2022) Clinical efficacy of platelet-rich plasma in the treatment of lateral epicondylitis: a systematic review and meta-analysis of randomized placebo-controlled clinical trials. Clin Rheumatol. 39(8):2255-2265.
- Chen XT, Fang W, Jones IA, Heckmann ND, Park C, et al. (2021) The Efficacy of Platelet-Rich Plasma for Improving Pain and Function in Lateral Epicondylitis: A Systematic Review and Meta-analysis with Risk-of-Bias Assessment. Arthroscopy. 37(9):2937-2952.
- Xu Y, Lin W, Qi Z, Yan Z, Yan, et al. (2026) Time-Dependent Efficacy and Safety of Percutaneous Treatments for Lateral Epicondylitis: A Systematic Review and Network Meta-Analysis. J Pain Res. 19:604185.
- Ye Z, Yuan Y, Kuang G, Qiu L, Tan X, Wen Z, et al. Platelet-rich plasma and corticosteroid injection for tendinopathy: a systematic review and meta-analysis. BMC Musculoskelet Disord. 2025;26(1):339.
- Chen X, Jones IA, Park C, Vangsness CT. (2018) The Efficacy of Platelet-Rich Plasma on Tendon and Ligament Healing: A Systematic Review and Meta-analysis With Bias Assessment. Am J Sports Med. 46(8):2020-2032.
- Chen X, Jones IA, Togashi R, Park C, Vangsness CT. (2020) Use of Platelet-Rich Plasma for the Improvement of Pain and Function in Rotator Cuff Tears: A Systematic Review and Meta-analysis With Bias Assessment. Am J Sports Med. 48(8):2028-2041.
- Dunivan Q, Allen M, Caplan BE, Cole BJ. (2026) Leukocyte-poor platelet-rich plasma reduces retear risk after arthroscopic rotator cuff repair: a meta-analysis with mechanistic and economic evaluation. J Shoulder Elbow Surg. 35(8):2037-2049.
- Hurley ET, Shimozono Y, Hannon CP, Smyth NA, Murawski CD, et al. (2020)Platelet-Rich Plasma Versus Corticosteroids for Plantar Fasciitis: A Systematic Review of Randomized Controlled Trials. Orthop J Sports Med. 8(4):2325967120915704.
- Ling Y, Wang S. (2018) Effects of platelet-rich plasma in the treatment of plantar fasciitis: A meta-analysis of randomized controlled trials. Medicine (Baltimore). 2018;97(37):e12110.
- Yang WY, Han YH, Cao XW, Pan JK, Zeng LF, et al. (2017) Platelet-rich plasma as a treatment for plantar fasciitis: A meta-analysis of randomized controlled trials. Medicine (Baltimore). 96(44):e8475.
- Martinez-Zapata MJ, Martí-Carvajal AJ, Solà I, Expósito JA, Bolíbar I, et al. (2016) Autologous platelet-rich plasma for treating chronic wounds. Cochrane Database Syst Rev. (5):CD006899.
- Del Pino-Sedeño T, Trujillo-Martín MM, Andia I, Aragón-Sánchez J, Herrera-Ramos E, et al. (2019) Platelet-rich plasma for the treatment of diabetic foot ulcers: A meta-analysis. Wound Repair Regen. 27(2):170-182.
- Hu Z, Qu S, Zhang J, Cao X, Wang P, et al. (2019) Efficacy and Safety of Platelet-Rich Plasma for Patients with Diabetic Ulcers: A Systematic Review and Meta-analysis. Adv Wound Care (New Rochelle). 8(7):298-308.
- Qu W, Wang Z, Hunt C, Morrow AS, Urtecho M, et al. (2021) The Effectiveness and Safety of Platelet-Rich Plasma for Chronic Wounds: A Systematic Review and Meta-analysis. Mayo Clin Proc. 96(9):2407-2417.
- Qu S, Hu Z, Zhang Y, Wang P, Li S, et al. (2022) Clinical Studies on Platelet-Rich Plasma Therapy for Chronic Cutaneous Ulcers: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Adv Wound Care (New Rochelle).11(2):56-69.
- Xia Y, Zhao J, Xie J, Lv Y, Cao DS. (2019) The Efficacy of Platelet-Rich Plasma Dressing for Chronic Nonhealing Ulcers: A Meta-Analysis of 15 Randomized Controlled Trials. Plast Reconstr Surg. 144(6):1463-1474.
- Hu Z, Wang S, Yang H, Xv H, Shan B, et al. (2024) Efficacy and safety of platelet-rich plasma in the treatment of venous ulcers: A systematic review and meta-analysis of randomized controlled trials. Int Wound J. 21(2):e14736.
- Guan X, Zhu G, Jiang H, Wang B, Cheng J. (2026) Comparative efficacy of advanced therapeutic modalities for diabetic foot ulcers: a systematic review and meta-analysis including NPWT, stem cells, ESWT, and bioengineered treatments. Front Bioeng Biotechnol. 14:1792670.
- Giordano S, Romeo M, di Summa P, Salval A, Lankinen P. (2018) A Meta-analysis On Evidence Of Platelet-rich Plasma for Androgenetic Alopecia. Int J Trichology. 10(1):1-10.
- Gupta AK, Cole J, Deutsch DP, Everts PA, Niedbalski RP, et al. (2019) Platelet-Rich Plasma as a Treatment for Androgenetic Alopecia. Dermatol Surg. 45(10):1262-1273.
- Anitua E, Padilla S, Prado R, Tierno R, Hamdan Alkhraisat M.(2026) Influence of platelet-rich plasma composition on pain and functional performance in knee osteoarthritis: a systematic review and network meta-analysis. Knee Surg Relat Res. X 38(1):17.
- Martín-Vega M, Gómez-Carrión Á, Zaragoza-García I, Ortuño-Soriano I, Marcelino-Argemi A, et al. Leukocyte-rich versus leukocyte-poor platelet-rich plasma for Osteoarthritis: A systematic review. Regen Ther. 31:101078.
- Lim JJ, Stepaniuk EA, Belk JW, Keeter C, Oeding JF, et al.(2026) Platelet Concentration Does Not Influence Clinical Efficacy and Retear Rates of Rotator Cuff Repair With Platelet-Rich Plasma: A Systematic Review and Meta-analysis With Meta-Regression. Am J Sports Med. 54(10):2597-2609.
- Hurley ET, Colasanti CA, Anil U, Luthringer TA, Alaia MJ, et al. (2021) The Effect of Platelet-Rich Plasma Leukocyte Concentration on Arthroscopic Rotator Cuff Repair: A Network Meta-analysis of Randomized Controlled Trials. Am J Sports Med. 49(9):2528-2535.
- Houck DA, Kraeutler MJ, Thornton LB, McCarty EC, Bravman JT, et al. (2019) Treatment of Lateral Epicondylitis With Autologous Blood, Platelet-Rich Plasma, or Corticosteroid Injections: A Systematic Review of Overlapping Meta-analyses. Orthop J Sports Med. 7(3):2325967119831052.
- Ali Elsiddig Ahmed E, Muharib R Alruwaili K, Alruwaili AH, Talal M Alruwaili A, Ahmed S Aljudia H, et al. (2025) Efficacy of Platelet-Rich Plasma in Treatment of Achilles Tendinopathy: Systematic Review and Meta-Analysis. Cureus. X 17(2):e79692.
- Gupta AK, Gupta AK, Bamimore M. (2022) Platelet-Rich Plasma Monotherapies for Androgenetic Alopecia: A Network Meta-Analysis and Meta-Regression Study. J Drugs Dermatol. X 21(9):943-952.
- Belk JW, Kraeutler MJ, Houck DA, Goodrich JA, Dragoo JL, et al. (2021) Platelet-Rich Plasma Versus Hyaluronic Acid for Knee Osteoarthritis: A Systematic Review and Meta-analysis of Randomized Controlled Trials. Am J Sports Med. 49(1):249-260.
- Belk JW, Lim JJ, Keeter C, McCulloch PC, Houck DA, et al. (2023) Patients With Knee Osteoarthritis Who Receive Platelet-Rich Plasma or Bone Marrow Aspirate Concentrate Injections Have Better Outcomes Than Patients Who Receive Hyaluronic Acid: Systematic Review and Meta-analysis. Arthroscopy. 39(7):1714-1734.
- Costa LAV, Lenza M, Irrgang JJ, Fu FH, Ferretti M. (2023) How Does Platelet-Rich Plasma Compare Clinically to Other Therapies in the Treatment of Knee Osteoarthritis? A Systematic Review and Meta-analysis. Am J Sports Med. 51(4):1074-1086.
- Hohmann E, Tetsworth K, Glatt V. (2020) Is platelet-rich plasma effective for the treatment of knee osteoarthritis? A systematic review and meta-analysis of level 1 and 2 randomized controlled trials. Eur J Orthop Surg Traumatol. 30(6):955-967.
- Anil U, Markus DH, Hurley ET, Manjunath AK, Alaia MJ, Campbell KA, et al. The efficacy of intra-articular injections in the treatment of knee osteoarthritis: A network meta-analysis of randomized controlled trials. Knee. 32:173-182.
- Migliorini F, Driessen A, Quack V, Sippel N, Cooper B, et al. (2021) Comparison between intra-articular infiltrations of placebo, steroids, hyaluronic and PRP for knee osteoarthritis: a Bayesian network meta-analysis. Arch Orthop Trauma Surg. X 141(9):1473-1490.
- Singh H, Knapik DM, Polce EM, Eikani CK, Bjornstad AH, et al. (2022) Relative Efficacy of Intra-articular Injections in the Treatment of Knee Osteoarthritis: A Systematic Review and Network Meta-analysis. Am J Sports Med. 50(11):3140-3148.
- Filardo G, Previtali D, Napoli F, Candrian C, Zaffagnini S, et al. (2021)PRP Injections for the Treatment of Knee Osteoarthritis: A Meta-Analysis of Randomized Controlled Trials. Cartilage. 2021;13(1_suppl):364S-375S.
- Chen Z, Wang C, You D, Zhao S, Zhu Z, et al. (2020) Platelet-rich plasma versus hyaluronic acid in the treatment of knee osteoarthritis: A meta-analysis. Medicine (Baltimore). 99(11):e19388.
- Hurley ET, Lim Fat D, Moran CJ, Mullett H. 2019)The Efficacy of Platelet-Rich Plasma and Platelet-Rich Fibrin in Arthroscopic Rotator Cuff Repair: A Meta-analysis of Randomized Controlled Trials. Am J Sports Med.47(3):753-761.
- Assi A, Hamyeh A, Mohanna R, Boutros M, Boueri W, et al. (2026) Platelet-rich plasma vs. placebo injections in achilles tendinopathy: A systematic review and meta-analysis of randomized controlled trials. J Foot Ankle Surg. X 65(4):139.e1-139.e5
- Chen YJ, Wu YC, Tu YK, Cheng JW, Tsai WC, et al. (2019) Autologous Blood-Derived Products Compared With Corticosteroids for Treatment of Plantar Fasciopathy: A Systematic Review and Meta-Analysis. Am J Phys Med Rehabil. 98(5):343-352.
- Xu Q, Chen J, Cheng L. (2019) Comparison of platelet rich plasma and corticosteroids in the management of lateral epicondylitis: A meta-analysis of randomized controlled trials. Int J Surg. 67:37-46.
- Antunes Júnior CR, Santos RSS, Barreto ESR, Azevedo GN, Lima EBS, et al. (2026) Platelet-Rich Plasma Does Not Improve Pain or Function in Patients With Lateral Epicondylitis as Compared With Placebo: A Meta-analysis of Randomized Clinical Trials. Am J Sports Med. 54(5):1247-1257.
- Evans AG, Mwangi JM, Pope RW, Ivanic MG, Botros MA, et al. (2022) Platelet-rich plasma as a therapy for androgenic alopecia: a systematic review and meta-analysis. J Dermatolog Treat. 33(1):498-511.
- Nakagawa HF, Kim J, Rabinowitz J, Sussman WI. (2026) Assessment of adverse events and safety associated with intra-articular platelet-rich plasma injections compared to other injectates for knee osteoarthritis: A systematic review and meta-analysis. PM R.
- Rayo-Martín R, Rayo-Rosado R, Rayo-Pérez AM, Sánchez-Morilla S, et al. (2025) Comparison of PRP Injections Versus Corticosteroid Injections in Plantar Fasciitis: Systematic Review and Meta-analysis. Foot Ankle Spec. 19386400251363002.

