The platelet-rich plasma and mesenchymal stem cell milieu: A review of therapeutic effects on bone healing

Abstract

Platelet-rich plasma is autologous plasma that contains concentrated platelets compared to whole blood. It is relatively inexpensive to produce, can be easily isolated from whole blood, and can be administered while the patient is in the operating room. Further, because platelet-rich plasma is an autologous therapy, there is minimal risk for adverse reactions to the patient. Platelet-rich plasma has been used to promote bone regeneration due to its abundance of concentrated growth factors that are essential to wound healing. In this review, we summarize the methods for producing platelet-rich plasma and the history of its use in bone regeneration. We also summarize the growth factor profiles derived from platelet-rich plasma, with emphasis on those factors that play a direct role in promoting bone repair within the local fracture environment. In addition, we discuss the potential advantages of combining platelet-rich plasma with mesenchymal stem cells, a multipotent cell type often obtained from bone marrow or fat, to improve craniofacial and long bone regeneration. We detail what is currently known about how platelet-rich plasma influences mesenchymal stem cells in vitro, and then highlight the clinical outcomes of administering platelet-rich plasma and mesenchymal stem cells as a combination therapy to promote bone regeneration in vivo.

1 |. CLINICAL NEED FOR BONE REGENERATION

Fractures, bony defects, and periodontal disease are common conditions that require bone regeneration in order to heal. Approximately 7.9 million bone fractures occur annually in the United States, with 5% to 10% resulting in delayed/nonunion.^1–3 Incidence of nonunion is increased for the elderly or those with comorbidities, such as osteoporosis and diabetes.^3,4

Bone healing is a series of orchestrated events. First, a hematoma forms around the injury site, which contains fracture debris that initiates an acute, pro-inflammatory cascade with M1 (classical) macrophages and other inflammatory cells.^5–7 Progression to an anti-inflammatory state, populated by M2 (alternatively activated) macrophages that secrete cytokines and growth factors for repair, is critical for healing.^6,7 Bone regenerates via two distinct pathways: intramembranous or endochondral ossification. Intramembranous ossification is direct bone formation and occurs when mesenchymal progenitor cells differentiate directly into osteoblasts, form trabeculated bone, and is then remodeled into cortical bone.⁷ Calvarial fractures tend to heal via intramembranous ossification.⁸ Endochondral ossification, which is the most common pathway through which long bone fractures heal, occurs when mesenchymal progenitors differentiate into chondrocytes, forming a cartilage template intermediate.^9–11 Chondrocytes within the fracture callus undergo hypertrophic maturation and then transform into osteoblasts and osteocytes, mineralizing and remodeling the cartilage matrix to bone.^12–14 This highly trabeculated bone is then remodeled to cortical bone in the final stage of healing.⁷

There is a clinical need to accelerate bone regeneration after a fracture, and therefore, numerous grafting materials have been tested in orthopaedics. Autogenous cancellous bone is currently the gold standard. However, it has well-described limitations, including limited volumes that can be harvested from donor sites, donor site pain, and prolonged healing time for the donor sites.^15,16 Other bone grafting materials, such as natural and synthetic polymers, ceramics, metals and composites, also have advantages and disadvantages. For example, synthetic polymers are biodegradable and versatile, but have low mechanical strength.¹⁷ Calcium-phosphate ceramics are generally biocompatible and osteoconductive, but are brittle compared to bone.¹⁷ These materials have been paired with orthobiologic therapies, which include stem cells, osteoinductive growth factors (such as those provided by platelet-rich plasma), osteoconductive matrices, and anabolic agents to enhance bone healing.¹⁸

Mesenchymal stem cells (MSCs) are a promising adjunctive therapy for fracture repair. MSCs are multipotent stromal cells that can differentiate into multiple cell types, including chondrocytes and osteoblasts, and play an important role in bone regeneration¹⁹ by providing a source of primary osteogenic cells, encouraging osteoinduction, and initiating osteoconduction.²⁰ Bone marrow-derived MSCs have been used to reduce healing time and aid in the repair of nonunion fractures.^19,21

2 |. PREPARING PLATELET-RICH PLASMA

Broadly, platelet-rich plasma (PRP) is autologous plasma that contains more platelets than whole blood. Most commonly, PRP is defined as plasma containing platelets at a concentration approximately two to five times higher than whole blood, depending on the preperation.^22–25 However, protocols can be adjusted to create PRP with higher or lower platelet concentrations. Not all PRP is reported as an increased platelet percentage over baseline. Marx defined PRP as a 5 mL volume of plasma containing 1 × 10⁶ platelets/μL because it enhanced bone and soft tissue healing.²⁵ While PRP is the most common nomenclature used, it is also referred to as platelet-rich concentrate, autologous platelet gel or plasma-rich growth factors.^25,26

Normal platelet counts in whole blood average 200 × 10³ to 250 × 10³ platelets/μL.^22,25 Platelets circulate in peripheral blood until they are activated to form a clot, which concentrates platelets to the injury site. The clotting cascade initiates healing via one of two pathways that converge in the latter steps resulting in platelet activation and release of proteins necessary for clot formation.²⁷ Following fracture, platelets are essential in the formation of a fracture hematoma, after which they degranulate and release growth factors from alpha granules that are crucial to the healing process.^2,25,28 Seventy percent of the stored growth factors are secreted within 10 minutes of clotting, and 90% to 95% within 1 hour.²⁵ Additional growth factors are synthesized and secreted for the remaining 7–8 days of the platelets’ lifespan.^25,26 While a naturally occurring hematoma is composed of 95% erythrocytes, a clot that has been formed using leukocyte-depleted PRP is composed of 95% platelets, 4% red blood cells, and 1% white blood cells.²⁷ While the effect of PRP is largely attributed to the increased concentration of growth factors necessary for wound healing, PRP is a complex, bioactive milieu with other factors, including plasma proteins, delta and lambda granules from platelets, and other cell types (leukocytes, erythrocytes) that may play a critical role in its biologic activity.²⁸

PRP production is not standardized and therefore, numerous PRP protocols have been reported with significant variations in anticoagulant, centrifugation technique and activation method used. For example, common anticoagulants used include ethylenediami-netetraacetic acid, acid citrate dextrose, and sodium citrate. Centrifugation protocols can include one (soft spin) or two (soft spin and hard spin) centrifugations with variable speed and times. Platelets can be activated endogenously via freeze-thaw cycles to disrupt the platelet membrane, however, exogenous platelet activation using either thrombin or calcium chloride is more common. Some authors caution against the use of thrombin because it can lead to the formation of peripheral blood clots and myocardial infarction.²⁹ Additionally, it has been reported that bovine thrombin could cross-react with human factor V leading to coagulopathies.²⁹ Lastly, commercial preparation kits have been shown to produce variable PRP. Oudelaar et al³⁰ reviewed 10 commercial PRP preparation kits and concluded that there was a large amount of heterogeneity between kits in regards to platelet, leukocyte and growth factor concentrations.

Another factor that contributes to PRP variability is the concentration of other cell types, such as leukocytes and erythrocytes. The centrifugation process used to produce PRP dramatically reduces or eliminates erythrocytes and therefore, erythrocyte content is often not reported. Nonetheless, as reviewed by Oryan et al³¹ erythrocytes should be excluded from PRP due to the potential for free radical production and induction of platelet aggregation. Further, under conditions of oxidative stress, heme (the protein-bound molecules that make up hemoglobin), can become free and cytotoxic.²⁸

Whether leukocytes should be retained in PRP is highly debated and the results testing leukocyte-rich versus pure (or leukocyte-poor) PRP are confounding. The majority of commercial PRP kits yield leukocyte-rich PRP in a review performed by Oudelaar et al.³⁰ This review also highlighted that there are both beneficial and adverse effects of leukocyte inclusion, and that leukocyte content should be matched to the specific clinical field and application. Regarding bone regeneration, leukocyte-rich PRP has been shown to stimulate osteogenic differentiation and proliferation of human MSCs in vitro in a dose-dependent manner compared to PRP without leukocytes.³² As reviewed by Oryan et al,³¹ high concentrations of leukocytes in PRP seem to retard bone healing by inducing an inflammatory reaction that may become chronic, whereas low concentrations of leukocytes may not be able to induce the necessary inflammatory response needed in early bone regeneration. Given these confounding reports, the effect of leukocyte content on bone healing is still unknown.³³

It is worthwhile to note that the confounding results of leukocyte content in PRP may be due to the fact that simple classification (rich vs poor) is inadequate. Leukocytes are classified as granulocytes (neutrophils, eosinophils, and basophils) or mononuclear cells (lymphocytes and monocytes/macrophages). The role of granulocytes, especially neutrophils, is to combat invading pathogens but they can also incite local tissue destruction.³⁴ Therefore, if the goal of PRP is to enhance healing, adding neutrophils in excess is likely antagonistic.²⁸ However, circulating monocytes (likely M2 phenotype) have been shown to suppress inflammation, promote angiogenesis, and support collagen synthesis through transforming growth factor β (TGFβ), vascular endothelial growth factor (VEGF), and fibroblast growth factor (FGF), respectively.²⁸ As these growth factors are also important for bone healing (discussed below), it is reasonable to hypothesize that the leukocyte cell subtype included in PRP may differentiate its impact on bone repair, while reporting this could improve our mechanistic understanding of PRP and its clinical efficacy.

3 |. BONE HEALING GROWTH FACTORS IN PRP

There are numerous growth factors in PRP that are important for wound healing: platelet derived growth factor (PDGFaa, PDGFbb, PDGFab), TGFβ₁/TGFβ₂, VEGF, FGF, epithelial growth factor (EGF), bone morphogenetic protein (BMP), hepatocyte growth factor (HGF) and insulin-like growth factor (IGF).^23,25,35 Upon platelet activation post-injury, growth factors are secreted through the platelet membrane and act upon target cells to stimulate cellular proliferation, vascular invasion (angiogenesis), matrix formation, osteoid production, and collagen synthesis.^23,25,26

Many of the growth factors within PRP play a direct role in promoting fracture repair. Activated PDGF attaches to transmembrane receptors on osteoblasts, osteoclasts, chondrocytes, fibro-blasts and macrophages to stimulate mitogenesis, angiogenesis, bone remodeling, and phagocytosis of damaged tissue during normal wound and fracture healing.^36,37 TGFβ has been shown to regulate proliferation, differentiation, chemotaxis and adhesion of progenitor cells in the wound bed.³⁸ Further, TGFβ is a potent chondroinductive growth factor that rapidly upregulates type II collagen expression in MSCs.³⁹ Chondrocytes and osteoblasts are enriched with TGFβ receptors supporting the idea that TGFβ plays a significant role in the bone healing process and the most intense TGFβ immunostaining occurs during chondrogenic proliferation and endochondral ossification.⁴⁰ VEGF, which is expressed by hypertrophic chondrocytes in the fracture callus, as well as the growth plate, plays a critical role in promoting vascular invasion into the avascular cartilage anlagen.^41,42 Further, VEGF acts synergistically with osteogenic proteins, such as BMP4 and BMP2, by enhancing cell recruitment, prolonging cell survival, increasing angiogenesis, accelerating cartilage resorption and enhancing bone mineralization.^43,44

4 |. PRP HISTORY IN ORTHOPAEDICS

Historically, PRP was used in cell culture beginning in the 1970s⁴⁵ and was first used in a clinical setting in 1987.⁴⁶ In the early 1990s, Marx used PRP for maxillofacial surgery,²⁶ which jumpstarted its use in bone repair. Today, PRP has been used as a bioactive agent to restore bone and soft tissue defects in orthopedic, maxillofacial and plastic surgery. PRPs use as a bioactive therapy has been reviewed extensively in bone, muscle, cartilage, tendon and ligament repair.^29,47–52 In this review, we will briefly discuss the clinical outcomes of PRP alone on bone healing and then focus on the effects of PRP combined with exogenous MSC therapy on intramembranous and endochondral bone regeneration.

5 |. CLINICAL OUTCOMES OF PRP ON BONE REPAIR

Most preclinical investigations show that PRP-treated bone defects exhibit more advanced healing than controls, however this finding is not universally consistent. Of the 29 PRP-treated animal long-bone studies included in the meta-analysis by Gianakos et al,⁵³ 16 of 18 (89%) showed statistically significant increases in bone healing and tissue differentiation rates while 9 of 11 (82%) showed qualitative improvement of bone regeneration. In studies that quantified healing, a statistically significant increase in the amount of consolidation, bone formation, and cortical bone thickness was reported. In qualitative studies, 7 of 9 (78%) reported improvement in bone consolidation.⁵³ Lastly, six studies investigated mechanical properties: five studies reported significantly higher torsional stiffness in the PRP-treated group while one study reported a significant increase in three-point load bearing resulting in increased bone strength after PRP treatment.⁵³ Based on these results, it is reasonable to hypothesize that PRP acts on the intrinsic stem cell niche to promote repair. However, while the majority of studies show a positive effect on bone healing, there are a minority that show no significant benefit.⁵³ Due to these conflicting published results, some investigators have suggested that PRP is not sufficient as a sole adjuvant to enhance bone regeneration.

6 |. IN VITRO OUTCOMES OF PRP + EXOGENOUS MSCs

Because of the osteogenic growth factors in PRP, it was hypothesized that PRP may act synergistically with MSCs to accelerate osteogenesis. Therefore, many studies have analyzed the effects of combining PRP and MSCs in vitro. As reviewed by Fernandes and Yang, PRP generally enhanced the proliferation and differentiation of MSCs in multiple species.²⁹ MSC proliferation was significantly increased with PRP present in the culture media when compared to DMEM controls with ITS + 1 (insulin, transferrin, selenium, linoleic-BSA) supplementation only.⁵⁴ In fact, cell proliferation showed a significant dose-dependent increase when MSCs were cultured with increasing concentrations of PRP (2%–10%).⁵⁵

In addition to increasing proliferation, PRP has also been shown to increase osteogenesis of MSCs in vitro.⁵⁶ Wei et al⁵⁷ reported that while PRP alone failed to induce osteogenesis of MSCs, adding PRP to osteogenic media dramatically increased mineralization. Further, when MSCs were combined with PRP in vitro, there was significantly increased alkaline phosphatase and osteocalcin synthesis indicating enhanced osteogenic differentiation.⁵⁸ Taken together, these in vitro studies demonstrate that PRP could improve the osteogenic potential of MSCs.

The effect of PRP on MSCs in vitro have also been investigated using scaffolds. MSCs had improved osteogenic differentiation when cultured with PRP and nanohydroxyapatite-type I collagen beads.⁵⁹ There was significantly increased formation of mineralized nodules as well as increased osteocalcin, collagen type 1 and 3, and tenomodulin gene expression in MSCs cultured with a PRP gel/calcium-phosphate (CaP) composite.⁶⁰ Further, in a small case study, when MSCs from nonunion patients were cultured in a collagen matrix/PRP complex, MSCs proliferated and differentiated into osteogenic progenitor cells within the clot.⁶¹

7 |. CLINICAL OUTCOMES OF PRP + EXOGENOUS MSCs IN CRANIOFACIAL APPLICATIONS

7.1 |. Bone healing outcomes in small animal models

PRP combined with MSCs has been studied extensively in periodontics, maxillofacial and calvarial defects that heal via intramembranous ossification (Table 1). Using a rabbit model, Hwang and Choi,⁶² studied the consolidation period in distraction osteogenesis of the mandible. The PRP + MSCs group had accelerated bone formation compared to the PRP and control groups, suggesting that the concentrated growth factors in PRP worked synergistically with MSCs to regenerate bone.⁶² PRP + MSCs in a sandwich-like scaffold resulted in significantly more bone formation in a critical-sized calvarial defect than the empty defect, PRP alone or MSCs alone.⁶³ Although reported platelet counts were lower than published normal ranges for rodents, bone filling in a calvarial defect was significantly greater when PRP was combined with adipose-derived stem cells (ASCs) compared to PRP alone or ASCs in a carrier scaffold.⁶⁴

TABLE 1.

PRP combined with MSCs for craniofacial healing

Reference	Species	Age	Sex	Sample size	Defect model	Cell type	PRP platelet concentration	Scaffold	Treatment groups	Major outcome
Hwang et al62	Rabbit	Adult	Male	38	Distraction osteogenesis of the mandible	Rabbit MSC	Not reported	PRP gel	Empty defect, PRP, PRP + MSC	Significantly more new bone in the distraction gap, thus reducing the consolidation period in the PRP + MSC group compared to PRP and empty defect.
Liu et al63	Rat	8 wk	Male	42	Critical-sized (8 mm), calvaria	Rat MSC ± genetically modified to express BMP2	Not reported	nCS	Empty defect, nCS, nCS + PRP, nCS +MSC, nCS +MSC + PRP, nCS +MSC/BMP2, nCS +MSC/BMP2 + PRP	Micro-CT and histology revealed nCS +MSC/BMP2 + PRP had significantly more bone formation than all other groups. Bone formation in nCS +MSC + PRP was not significantly different from nCS +MSC/BMP2 but was significantly greater than remaining groups.
Tajima et al64	Rat	11 wk	Male	50	Calvaria (5 mm)	Rat ASC	180 × 104/mL	PRP gel or type I collagen gel	PBS, collagen, PRP, ASC + collagen, ASC + PRP	Newly formed bone was significantly greater in the ASC + PRP group compared to all other groups at 4 and 8 wk after transplantation.
Yamada et al65	Dog	Adult	Not reported	4	Mandible (10 mm)	Dog MSC	1.3 × 106 (no unit given)	PRP gel or PCBM	Empty defect, PCBM, PRP, PRP + MSC	PRP + MSC can elicit bone regeneration equivalent to PCBM and significantly more than PRP and empty defect.
Yamada et al66	Human	53–74 y	Male and female	3	Alveolar ridge atrophy	Human MSC	Not reported	PRP gel	MSC + PRP	Treatment with MSC + PRP resulted in complete coverage of the implant, absence of mobility at 6 mo, and marginal bone resorption.
Yamada et al67	Human	43–74 y	Male and female	8	Severe bone resorption of the alveolar arrest in the maxilla	Human MSC	Not reported	PRP gel	MSC + PRP	All implants were clinically stable, and the prostheses functioned during the follow up period. Alveolar bone height significantly increased compared to pre-operative bone height.
Yamada et al68	Human	19–78 y	Male and female	104	Dental cases requiring bone regeneration	Human MSC	Not reported	PRP gel	MSC + PRP	Post-operative bone density was significantly greater than pre-operation.
Tobita et al69	Dog	8–10 mo	Not reported	8	Class III bifurcation defect at P2, P3, P4	Dog ASC	946 × 103/mm3 (3-fold increase over whole blood)	PRP gel	Empty defect, PRP, PRP + ASC	PRP + ASC had more new bone formation but not significantly different from PRP alone.
Yun et al70	Dog	1 y	Male	4	Three-wall intrabony (periodontic, 4 × 4 × 4 m-m)	Human MSC	1×106/μL	HA	HA, HA + MSC, HA + PRP, HA + MSC + PRP	Bone density and bone-to-implant contact were greatest in the HA + MSC + PRP group but not significantly different from the other groups.

Abbreviations: ASC, adipose-derived stem cell; BMP, bone morphogenetic protein; CT, computed tomography; HA, hydroxyapatite; MSC, mesenchymal stem cell; nCS, nano-calcium sulfate; PCBM, particulate cancellous bone and marrow; PRP, platelet-rich plasma.

7.2 |. Bone healing outcomes in large animal models

Dogs are a common model in periodontal and maxillofacial research (Table 1). Yamada et al^65–68 reported defects filled with PRP + MSCs were equivalent to particulate cancellous bone and marrow (PCBM; autogenous bone) and superior to MSCs alone and empty defects, resulting in new bone formation after 2 weeks and a tubular pattern with abundant vascularization at 8 weeks. However, in another canine periodontal defect model, the PRP + ASCs co-implantation group had more new bone formation (64%) after 2 months, but was not significantly different from PRP alone (54%).⁶⁹ Further, a pilot study by Yun et al⁷⁰ reported that PRP + MSCs in a hydroxyapatite (HA) scaffold had increased bone density (62% at 6 weeks, 72% at 12 weeks) and increased bone-to-implant contact (22% at 6 weeks, 42% at 12 weeks) in adult dogs with three-wall intrabony defects, a periodontal defect, compared to PRP + HA, MSCs + HA, or HA alone, but it failed to reach statistical significance.

7.3 |. Bone healing outcomes in human patients

Human PRP + MSCs studies are mostly limited to dentistry and maxillofacial surgery (Table 1). Titanium implants surrounded by coagulated PRP + MSCs have shown some success as indicated by complete coverage of the implant, absence of mobility at 6 months, and marginal bone resorption in three patients with severe maxillary alveolar ridge atrophy.⁶⁶ Yamada et al⁶⁷ reported the average alveolar bone height was significantly increased at 3 and 6 months after osteotome sinus floor elevation combined with PRP + MSCs compared to pre-operative bone height. Further, bone density was significantly increased at all time points post-operation compared to pre-operation when PRP + MSCs were transplanted in 104 dental cases requiring bone regeneration.⁶⁸ Combining PRP with MSCs appears to be a promising therapy for craniofacial applications but controlled studies are lacking.

8 |. CLINICAL OUTCOMES OF PRP + EXOGENOUS MSCs IN LONG BONE APPLICATIONS

8.1 |. Bone healing outcomes in small animal models

Rodent models have been used extensively to study the outcome of PRP + MSCs on long bone healing (Table 2). Wei et al⁵⁷ demonstrated a positive effect on tibial bone healing in ovariectomized (OVX) rats. By day 14 post-injury, no bone callus had formed in the OVX group, a thin callus formed in the OVX + PRP and OVX + MSCs groups, and a thick callus formed in the OVX + PRP + MSCs and non-OVX groups. The calluses had completely mineralized by 42 days in the OVX + PRP + MSCs and non-OVX groups and lamellar bone had fully formed in the OVX + PRP + MSCs rats whereas woven bone remained in the OVX, OVX + PRP and OVX + MSCs groups.⁵⁷ PRP + MSCs has demonstrated favorable results in distraction osteogenesis as well. Kawasumi et al⁷¹ utilized a rat limb lengthening model to demonstrate that callus formation was enhanced in a platelet dose-dependent manner within MSC scaffolds. At 4 weeks post-distraction, the high concentration group (1055% platelet concentration of whole blood) was the only group with radiological union and had significantly larger areas of bone mineralization compared to low (117%) and medium (352%) platelet concentrations.⁷¹ Femoral defects treated with PRP + MSCs gel combined with a CaP composite regenerated dense, cortical bone tissue that was indistinguishable from normal bone in 12-week-old rats.⁶⁰ However, when PRP + MSCs gel was administered alone, it resulted in thin, woven bone tissue regeneration that was superior to controls, but was still distinguishable from normal bone at 4 weeks post-operation.⁶⁰

TABLE 2.

PRP combined with MSCs for long bone healing

Reference	Species	Age	Sex	Sample size	Defect model	Cell type	PRP platelet concentration	Scaffold	Treatment groups	Major outcome
Qi et al60	Rat	12 wk	Male	18	Femur (2.5 × 5 mm)	Rat MSC	13.2 × 108/mL (6-fold increase from serum)	PRP gel ± CaP	Empty defect, MSC + PRP, CaP, PRP + CaP, MSC + CaP, MSC + PRP + CaP	MSC + PRP + CaP regenerated significantly more bone tissue at 4 wk than other groups.
Wei et al57	Rat	Adult	Female	100	Ovariectomized, tibia (1.5 mm)	Rat MSC	Not reported	PRP gel	Non-OVX, OVX, OVX + MSC, OVX + PRP, OVX + PRP + MSC	PRP + MSC aided bone callus mineralization by day 42 and lamellar bone formed in OVX + PRP + MSC and non-OVX groups.
Kawasumi et al71	Rat	9 wk	Male	91	Limb lengthening, femur	Rat MSC	High = 4358 × 103/μL; Med = 1453 × 105/μL; Low = 48 × 106/μL	PRP gel or collagen gel	MSC + PPP, MSC + PRP low, MSC + PRP med, MSC + PRP high, MSC + collagen gel	Significantly larger area of mineralized bone in MSC + PRP high group at 4 wk. No significant difference among other groups.
Lin et al59	Rabbit	12–16 wk	Male	36	Femoral epicondyle (0.5 × 1.0 cm)	Rabbit MSC	105 × 107/mL (3.7-fold increase over whole blood)	Nanohydroxyapatite-CIB or PRP gel	Empty defect, drilling treatment, PRP, PRP + MSC, CIB + PRP, CIB + PRP + MSC	CIB + PRP + MSC had enhanced osteogenesis and accelerated mineralization compared to other groups.
El Backly et al72	Rabbit	Not reported	Male	10	Ulna (1.4 cm)	Rabbit MSC	3 × 106/μL	PRP + MSC gel with nanohydroxyapatite/poly (ester urethane) scaffold	Scaffold+PBS, scaffold+MSC, scaffold+MSC + PRP	Micro-CT quantification revealed MSC + PRP had twice as much bone regeneration as MSC or scaffold alone.
Park et al73	Rabbit	Adult	Male	30	Femur (2 cm)	Rabbit MSC	Not reported	Bone graft	PRP + bone graft, PRP + MSC + bone graft	MSC + PRP enhanced bone formation and increased growth factor production in grafts.
Kasten et al74	Rabbit	6–9 mo	Female	36	Critical-sized (15 mm), Radius	Rabbit MSC	10 × 108/mL	CDHA	Empty defect, defect + autogenous bone, CDHA, CDHA + MSC, CDHA + PRP, CDHA + MSC + PRP	PRP and MSC increased bone regeneration individually compared to empty defect, but no additional effect of PRP + MSC combined.
Yu et al58	Rabbit	10 mo	Male	24	Radius (12 mm)	Rabbit MSC (osteogenic)	1056 × 103/μL (5.3-fold increase over whole blood)	bTCP	MSC + bTCP, PRP + MSC + bTCP	PRP + MSC enhanced osseous callus and increased radiograph score at 8 wk.
Lucarelli et al75	Sheep	3–5 y	Female	10	Critical-sized (3 cm), Metatarsus	Ovine MSC	1 × 106/mL	PRP gel + collagen matrix, Bone allograft	PRP + MSC + bone graft, bone graft	Increased vascular invasion, more bone regeneration, and higher extraction torque test values in PRP + MSC + bone allograft group.
Qiu et al76	Minipigs	12–18 mo	Female	12	Femoral condyle (8 × 10mm)	Minipig MSC	1.7 × 109/mL (4-fold increase over whole blood)	CPC	CPC, CPC + MSC, CPC + MSC + PRP	New bone formation and blood vessel density was: CPC + MSC + PRP > CPC + MSC > CPC, all significantly different from each other.
Nair et al77	Goat	Adult	Not reported	6	Femur (2 cm)	Goat MSC (osteogenic)	6 × 108/mL (1.5-fold increase over whole blood)	HASi	HASi, HASi+MSC, HASi+MSC + PRP	No difference in new bone formation or material degradation between the three groups.
Niemeyer et al78	Sheep	3 y	Female	20	Tibia (3 cm)	Ovine MSC, ovine ASC	1 × 109/mL (4–5-fold increase over whole blood)	Collagen sponge	Collagen, MSC + collagen, ASC + collagen, PRP + ASC + collagen	MSC had significantly more bone formation than collagen and ASC groups. PRP + ASC had more bone formation but lacked statistical significance.
Kitoh et al79	Human	12–20 y	Male and female	46	Distraction osteogenesis, femur and tibia	Human MSC	26.7 × 105/μL	PRP gel	No cellular treatment, MSC + PRP	Patients with MSC + PRP had significantly less time until osseous consolidation and fewer complications than patients without MSC.
Wittig et al61	Human	27–81 y	Male and female	3	Nonunion fractures; tibia, fibula, femur	Human MSC	Not reported	Collagen sponge	MSC + PRP + collagen sponge	All nonunion fractures had healed within 1–3 y after MSC + PRP treatment.

Abbreviations: ASC, adipose-derived stem cell; bTCP, β-tricalcium phosphate; CaP, calcium phosphate composite; CDHA, calcium-deficient hydroxyapatite ceramic; CIB, type I collagen bead; CPC, calcium phosphate cement; HASi, triphasic ceramic-coated hydroxyapatite; MSC, mesenchymal stem cell; OVX, ovariectomized; PRP, platelet-rich plasma.

Rabbit is another common model for studying PRP + MSCs therapy for long bone regeneration (Table 2). Femur defects filled with nanohydroxyapatite-type I collagen beads combined with PRP + MSCs had improved osteogenesis via accelerated mineralization compared to PRP + collagen beads or PRP alone.⁵⁹ El Backly et al⁷² combined PRP + MSC in a periosteal engineered membrane wrapped around an ulnar defect and demonstrated enhanced bone filling compared to empty and MSC-only scaffold controls. Further, treatment with PRP + MSCs in a femoral segmental bone defect resulted in enhanced bone formation and gene expression of growth factors in grafted tissues compared to treatment with PRP alone.⁷³ In a study using PRP and osteogenic-induced MSCs loaded onto a beta-tricalcium phosphate (bTCP) scaffold, Yu et al⁵⁸ tested bone regeneration following a segmental radial defect. At 8 weeks post-operation, the PRP + MSCs + bTCP group had enhanced osseous callus around the periphery of the implant and significantly higher radiograph scores than the MSCs+bTCP group without PRP. Although most studies report a synergistic response when PRP is combined with MSCs, Kasten et al⁷⁴ reported that PRP and MSCs both increased bone regeneration individually, compared to empty defects, however, there was no additional effect on bone healing when PRP and MSCs were combined.

8.2 |. Bone healing outcomes in large animal models

Studies utilizing PRP + MSCs therapy in large animal models are more limited (Table 2). Lucarelli et al⁷⁵ reported increased new bone inside allografts in the PRP + MSCs group compared to control, with less space between the host bone and the graft on radiographs, clear signs of ossification with presence of mature lamellar bone, and 4–5 times the vessel penetration into the graft from the host bone in ovine metatarsal critical sized defects. In another study, PRP + MSCs enhanced vascularization and new bone formation. At 12 weeks, calcium phosphate cement (CPC) scaffold combined with PRP + MSCs had significantly more new bone formation in the femoral condyle of minipigs than both the CPC + MSCs and CPC groups.⁷⁶ Histology confirmed new bone was being deposited by osteoblasts while osteocytes and new blood vessels were surrounded by woven bone matrix.⁷⁶ Further, the CPC + PRP + MSCs group had significantly greater blood vessel density than the CPC + MSCs group and was two-fold that of the CPC control.⁷⁶

However, there is conflicting evidence that PRP + MSCs may not enhance bone regeneration. In a small study using 6 adult goats, there was no significant difference in bone regeneration between PRP + osteogenically induced MSCs and osteogenically induced MSCs alone in a segmental femoral defect at 2 months post-operation.⁷⁷ In another study, roughly 25% of the defect was filled with new bone when PRP + ASCs were administered to a mid-diaphyseal tibial defect in sheep compared to ASCs alone (~10% filled with new bone), but failed to reach statistical significance.⁷⁸

8.3 |. Bone healing outcomes in human patients

PRP + MSCs therapy studies to treat long bone defects are extremely limited in humans (Table 2). Regarding distraction osteogenesis of human long bones, 16 patients that were treated with PRP + MSCs at the bone lengthening sites experienced significantly less time (34 days vs 73 days) until osseous consolidation and fewer complications (6% vs 23%) than the 30 patients treated with no additional cell therapy.⁷⁹ In another small case study, three patients with nonunion femur, tibia and/or fibula fractures were administered autologous PRP + MSCs with collagen matrix at the defect site.⁶¹ Within 3–6 months, radiographs revealed all patients had radiopaque (osteogenic) areas at the site of PRP + MSCs implantation and the nonunion fractures had completely healed within 1–3 years.⁶¹ Although these results are favorable for long bone healing in human patients, controlled studies utilizing PRP and MSCs are lacking.

9 |. REGULATORY ENVIRONMENT IN THE UNITED STATES

The use of human cells and tissue products are regulated by the US Food and Drug Administration (FDA) consisting of three tiers; low: no manipulation, middle: minimal manipulation, and high: beyond minimal manipulation. Low-tier products such as PRP and non-manipulated MSCs are exempt from FDA regulations, whereas middle-tier products are subject to infection and contamination prevention requirements and high-tier products are subject to the same pre-marketing requirements as any other drug or device.⁸⁰ Therefore, it is important to emphasize that most of the pre-clinical studies cited in this review utilize manipulated cells (e.g. culture-expansion, osteogenic induction) whereas the typical point-of-care clinical applications utilize cells without manipulation.

10 |. CONCLUSIONS

PRP shows promise as a biologic adjunct to promote bone repair, particularly when used in combination with MSCs. However, when the body of scientific literature is reviewed as a whole, it is evident that PRP with or without MSCs is not a proven therapeutic. The optimal method for preparation, final platelet concentrations, activation processes, inclusion/exclusion of other cell types, dose and dose intervals have yet to be determined and optimization may differ between species, donors, and tissue application. Further research is required to understand how platelet, growth factor and specific leukocyte subtype concentrations within PRP can be customized to promote bone repair in specific clinical applications. The balance of data suggests that a combination of PRP + MSCs appears to be superior for bone healing as compared to PRP alone. However, lack of PRP standardization and the MSCs tissue source have not been consistent, thereby introducing a great deal of variability in outcomes and conclusions. As an additional confounder, investigators have utilized varying bone healing models (intramembranous versus endochondral ossification) and it is likely that the influence of PRP with and without MSCs may differ in efficacy in each of these bone healing pathways. Therefore, there is a critical need for more controlled studies to elucidate the mechanism of repair when PRP is combined with MSCs in various types of bone healing environments.

11 |. FUTURE DIRECTIONS

PRP production and reporting is not currently standardized. Therefore, PRP can vary widely due to interpatient and intrapatient variability. The most common factors that introduce interpatient variability include the centrifugation protocol/commercial preparation kit used and the inclusion/exclusion of white blood cells. In a systematic review of PRP and long bone healing, Gianakos et al⁵³ reported that only 55% of studies reported the platelet count or concentration in their PRP preparations and no studies reported the white blood cell count. This lack of consistency makes it hard to draw conclusions on bone healing between different modalities. Therefore, it is necessary to advocate for consistency in reporting methods, such as platelet concentration, volume, and dose interval.⁸¹ DeLong et al³³ developed a PRP classification system called PAW which is based on the Platelet concentration, Activation method used, and presence or absence of White cells in the PRP. A more comprehensive classification system called Minimum Information for Studies Evaluating Biologics in Orthopaedics (MIBO) was developed for minimum reporting requirements for clinical studies evaluating PRP.⁸² Twenty-three experts compiled a 25-item checklist including details regarding study design, recipient details, injury details, intervention, whole blood processing, whole blood characteristics, PRP processing, PRP characteristics, activation, delivery, postoperative care, and outcome that should be reported for each study.⁸² Adoption of these systems would encourage standardization, may provide clarity regarding the effects of PRP on bone healing, and could lead to the development of optimized PRP protocols tailored to specific tissue types and pathological process. The ideal platelet concentration, volume per PRP treatment, number of treatments, and timing of injections for optimal bone regeneration remains to be determined.

It is also important to determine the optimal scaffold to deliver PRP to bone defects. Fibrin gel is a common scaffold, but additional scaffolds, such as hydroxyapatite, beta-tricalcium phosphate and calcium phosphate cement, are also used and could potentially alter PRP properties, such as growth factor release. A review of PRP combined with MSCs utilizing different scaffolds for bone defects is warranted.

Studies utilizing PRP supplemented with gene therapy are limited but genetically modified MSCs have been combined with PRP to enhance bone regeneration therapies. Several approaches using MSCs genetically modified to express BMP2,⁶³ angiopoietin-1,⁸³ and VEGF⁸⁴ have shown enhanced bone regeneration when combined with PRP.

Lastly, it is important to consider the bone lesion chronicity when designing studies. Most studies using long bone animal models surgically create acute bone defects that are immediately treated with PRP plus or minus MSCs prior to closure.⁵³ However, in clinical practice, delayed union is a chronic condition.⁵³ Therefore, it would be ideal if future in vivo studies examined the effects of PRP with or without MSCs on chronic bone pathologies.