Experimental study of negative pressure wound therapy combined with platelet-rich fibrin for bone-exposed wounds

Wounds with exposed bone are refractory, and are usually caused by mechanical trauma, burns, infections, pressure ulcers, diabetic illness and tumor resection [1,2]. Surgery is the most effective method for covering this type of wound, and local or distal skin flaps are typically needed to repair the defect [3]. However, due to potential donor site injury, there is increasing interest in identifying alterative treatment strategies. Skin grafting may be a simpler approach but skin grafting on exposed bone is contraindicated, as the avascular interface of the bone renders granulation tissue difficult to grow and the condition occasionally worsens when the periosteum is striped or corroded.

Recently, negative pressure wound therapy (NPWT) has emerged as a treatment option for complex wounds [4]. This therapy delivers sub atmospheric pressure through a porous foam dressing, covered with an adhesive drape. NPWT can remove excessive exudate and improve tissue blood circulation [5], reducing tissue edema and bacteria load [6]. Generally, NPWT maintains a closed and stable environment that is favorable for granulation tissue growth [7,8]. A few studies that utilized NPWT to treat bone-exposed wounds in rodent models have shown promising results [2,8]. However, the healing in these studies may be mainly attributed to wound contraction, rather than re-epithelialization or granulation tissue growth [9]. This is consistent with clinical observations, as applying NPWT to cover bone-exposed wounds does not seem to work well [2]. In previous studies, NPWT was usually combined with an autografted or allografted dermal substitute [10,11]. We believe that the key factor inhibiting granulation tissue growth in bone-exposed wounds is the inability of granulation tissue to grow vertically from the bone interface, as it must invade from the periphery to the center in the horizontal direction. This inefficient manner of growth is more notable with larger wounds. Therefore, we hypothesized that the optimal solution in promoting granulation in bone-exposed wounds is to improve the migration ability of inflammatory cells. In the present study, we used platelet-rich fibrin (PRF) in combination with NPWT. PRF is an autologous blood-derived product that contains a variety of cytokines and provides fibronectin scaffolds [12] directly leading the migration of inflammatory cells [13,14], which is different from the first autologous blood-derived product, platelet-derived plasma – platelet-rich plasma (PRP). The preparation method for PRF is easier than PRP as it achieves the gel without any manipulation of the blood such as anticoagulants, compared with PRP. With the technique of the PRF, the production was greater simplicity for the absence of manipulation that leads to a reduced possibility of alteration of the protocol due to an error of the operator [15]. A skull-exposure murine model was used to determine whether this combination strategy could achieve mesenchymal healing with granulation tissue.

Materials & methods

Animals

In total, 32 healthy Sprague-Dawley (SD) rats (male, weight 250–300 g, 10–12 weeks old) were used in this study. All rats were purchased from the Experimental Animals Centre of Fujian Medical University. All rats were treated in accordance with animal protocols approved by the ethics committee at Fujian Medical University. All rats were housed at the Animals Centre of Fujian Medical University.

Bone-exposed wound model

On the day before surgery, the hair was removed from the scalp of the rat using depilatory cream to maintain a hairless state [16–18]. Surgery was performed under general anesthesia (60 mg/kg pentobarbital) [19]. A circular wound, 1.5 cm in diameter, was made on the anterior skull area, and all tissues (including the periosteum) were excised from the surface of the skull. The wound edge was fixed with an adhesive frame (DuoDERM^® CGF^®; ConvaTec Ltd, NJ, USA) [2]. The wounds were sequentially covered with a white polyurethane foam dressing (vacuum sealing drainage [VSD]; WEGO Ltd, Shandong, China) and a semi-occlusive dressing (3 M™ TegadermTM™; 3M Science, MN, USA). Wounds receiving NPWT were then continuously connected to a vacuum system at -125 mmHg [19] through a tube (Figure 1).

PRF preparation

To avoid immunological rejection, autogenetic blood was used to prepare PRF. After ventricle centesis, blood (3.5 ml) was withdrawn from the rat’s heart, then immediately transferred to a centrifuge tube and centrifuged at 600 g for 10 min [20]. After centrifugation, the clot in the middle layer comprised PRF with abundant growth factors.

Study groups

To evaluate the treatment effects in bone-exposed wounds, the rats were randomly divided into four groups. In control group, normal saline (1.5 ml) was injected into the wounds. In NPWT group, the wound was covered with VSD and connected to a NPWT device at -125 mmHg. In PRF group, the wounds were covered with PRF alone. In NPWT + PRF group (N + P group), the wounds were covered with PRF, followed by NPWT under a negative pressure of 125 mmHg. To avoid occlusion of the NPWT caused by the gel, the PRF should be applied to the entire wound surface in a thin layer and the NPWT dressing covered it a littler wider. All wounds were closed with a 3 M membrane, and dressings were changed on days 4 and 7.

Wound closure analysis

A portable studio was used to harvest standardized photos of the wound on days 4, 7 and 10. The bone exposure area, wound contraction and granulation tissue growth were macroscopically quantified using Image J software, based on standardized photos.

Histological analysis

On days 7 and 10, granulation and skin tissues were completely excised along the surface of the skull. Tissues were stored in 4% paraformaldehyde for 24 h; paraffin-embedded sections were then prepared. Hematoxylin-eosin (HE) staining and Masson staining were used to characterize angiogenesis and collagen morphology, respectively. Finally, Image J software was used for the quantification of microvessel density (MVD), based on sections.

ELISA

On day 10, the wounds were sampled and homogenized in a RIPA lysate buffer (Beyotime lns, Shanghai, China). Supernatants of the wound homogenates were submitted to ELISAs for the assessment of the levels of four growth factors, VEGF, bFGF, PDGF-BB and TGF-β (R&D Systems Ltd, MN, USA). Briefly, the supernatant of the sample was added to a 96-well polystyrene microplate, which was pre-coated with a monoclonal antibody specific for growth factors and after antigen-antibody binding, the enzyme-linked antibody was used to clamp the objective protein. Finally, a colorimetric substrate was added for quantification.

Statistical analysis

All continuous data are expressed as the mean ± standard deviation. Statistical analyses were performed using GraphPad Prism 7.0 software (GraphPad Software, CA, USA). Comparisons among groups were performed by random-effects two-way analysis of variance. Significance was established as p < 0.05.

Results

Wound healing: NPWT promoted wound contraction, while PRF stimulated granulation tissue growth

All rats tolerated the surgery well and all NPWT devices were well fixed during the experiment. Due to the use of semi-occlusive dressing, no wounds showed indications of infection. All wounds showed a sequential healing process, including wound contraction and granulation tissue growth. Although the PRF remained in place, it shrank toward the center of the wound over time. On day 10, which was set as the end point of the experiment, the bone-exposed wounds of the N + P group were completely covered by granulation tissue (Figure 2). During the entire experiment, the proportion of the bone-exposed area was significantly smaller in the three experimental groups than in the control (p < 0.05). Additionally, the proportion of the bone-exposed area was smaller in the N + P group than in the PRF and NPWT groups, p < 0.05 (Figure 3A & B). Further analysis of the contributions of wound contraction and granulation tissue growth showed that, at all time points, wound contraction was significantly greater in the NPWT and N + P groups than in the PRF and control groups, p < 0.05 (Figure 3C). On day 4, the granulation tissue-covered area was significantly greater in the PRF and N + P groups than in the other two groups (p < 0.05). On days 7 and 10, the granulation tissue-covered area was greater in the PRF group than in the other three groups (p < 0.05), while the granulation tissue-covered area in the N + P group had decreased.

NPWT combined with PRF treatment increased angiogenesis & collagen formation

There was a large amount of inflammatory cells infiltrating the wound on day 4; thus, the results are not shown. On HE staining, the tissue of control remained infiltrated by inflammatory cells (Figure 4A, black arrows) on day 7 and there few vessel-like structures (Figure 4A, asterisks). There were more vessel-like structures in three experimental groups than in control, and red blood cells were observed in the structures in PRF and N + P groups. On day 10, all groups had obvious vessel-like structures; however, in control only, red blood cells were not observed. On Masson staining, fibroblasts were observed in three experimental groups, arranged directionality to an extent, with more consistent directionality in NPWT and N + P groups. In addition, collagen fibrin was plentiful and uniform in N + P group than in the other three groups (Figure 4A, white arrows). The MVD analysis revealed that the MVD had increased by about 20.88 ± 0.27% in N + P group, which was appropriately twice as much as that in NPWT group and PRF group (12.13 ± 0.25% and 12.74 ± 0.51%, respectively). In contrast, the proportion of shaped blood vessels in control was only 4.657 ± 0.16%. From a statistical point of view, the MVD was significantly higher in N + P group than in the other three groups (p < 0.05). Additionally, the MVD was significantly higher in NPWT and PRF groups than in control (p < 0.05). It should be noted that there was no statistical difference between NPWT group and PRF group in the MVD (Figure 4B).

NPWT combined with PRF therapy promoted growth factor upregulation

The ELISA results revealed that the levels of all four growth factors (VEGF, PDGF-BB, TGF-β and bFGF) in granulation tissue were significantly higher in N + P group than in the other three groups (p < 0.01). Additionally, all four growth factor had significantly higher levels in PRF group than in control (p < 0.05; Figure 5). While TGF-β and bFGF levels were significantly higher in NPWT group than in control (p < 0.05; Figure 5C & D) PDGF-BB and TGF-β levels were significantly higher in PRF group than in NPWT group (Figure 5A & D). Though bFGF levels in were higher in NPWT group than PRF group, it was not significant (Figure 5C).

Discussion

Repairing bone-exposed wounds is one of the greatest clinical challenges that plastic surgeons must face [21], especially when the bone surface is without the periosteum. This is because tissue repair cells, including inflammatory cells, fibroblasts and vascular endothelial cells, cannot proliferate on the avascular cell-repellent interface [22], which eventually leads to chronic wounds. Traditionally, tissue flaps have been used as the major therapy in repairing bone-exposed wounds. However, flap transplantation is limited by many conditions [22] and inevitably causes defects at the donor site. Recently, artificial materials, such as Pelnac™ [23], in combination with primary or secondary skin grafting have been gradually applied in clinical practice; however, the extra cost places a burden on the patients. Here, we proposed the use of NPWT combined with PRF in the treatment of bone-exposed wounds. The results of the present study showed that this combination therapy accelerated both granulation growth and wound contraction, which would greatly shorten the wound healing time. Wound healing has been traditionally divided into three stages: inflammation, proliferation and remodeling. Epidermalization, angiogenesis, granulated tissue formation and collagen deposition constitute the principal steps of the proliferation phase of wounds healing [24]. Wound healing is classically characterized as the transient development of the granulation tissue, which supports the rapid proliferation, migration and differentiation of adjacent epithelial cells. The restore of lost or damaged skin structures involves many processes, in which conjugated polysaccharide-protein and collagen deposit in wound tissue as the main ingredients [25]. The slow growth of granulated tissue and the reduction of collagen deposition will lead to chronically unhealed wounds, where epithelial cells located at the edge of the skin begin to proliferating and emit projections to re-establish the protective barrier to prevent body fluid losses and further bacterial invasion, called ‘re-epithelialization’ process [24].

The closure of the ordinary wound is mainly through the combined effect of re-epithelialization and wound contraction. Rodent models of excisional wound healing have been widely used to study each phase of healing; however, rodent healing models are generally considered limited because it is believed that rodent wounds heal mainly through contraction, while humans do re-epithelize to heal. In fact, in the rodent healing model, the early wound closure mainly relies on re-epithelialization, and it is not until the day 20 that the contribution of wound contraction to wound closure increases significantly [26]. As mentioned above, the epithelialization process of wounds cannot be separated from the support of granulation tissue [24]. In the bone-exposed wound model, the growth of granulation tissue becomes slow and difficult, due to the lack of periosteum. Therefore, promoting the formation of granulation tissue is particularly important for accelerating wound healing.

As the largest organ of the human body, skin and the underlying subcutaneous tissue are constantly subjected to external and internal mechanical forces. The extrinsic forces such as gravity, physical movement and trauma, as well as intrinsic forces generate by blood flow or skeletal growth. Endothelial cells, keratinocytes and fibroblasts (among others) can respond to these mechanical forces. Given the existence of cellular mechanical sensitivity of these cells, the spatial and temporal response of the skin to mechanical stimulation attract increasing interest in wound management today. Increasing clinical demand, coupled with a better understanding of the mechanical conduction properties of the skin, has driven major innovations in wound healing therapies that based on mechanical forces [7].

The production of robust granulation tissue during NPWT treatment has been reported in numerous clinical studies [4,27,28]. NPWT with a porous foam dressing evenly distribute vacuum to the wound, while causing microdeformation, have been referred to as microdeformation wound therapy (MDWT) [6,7]. Four primary mechanisms have been proposed: wound shrinkage or microdeformation [19,29]; microdeformation at the foam-wound surface interface [30]; fluid removal [29]; and stabilization of the wound environment [19]. Furthermore, several secondary effects such as, angiogenesis [19], granulation tissue formation [31], cellular proliferation, differentiation and migration [19,30], might likely involved in mechanotransduction pathways that alter the biology of wound healing.

Although NPWT is considered multifunctional, the major mechanism of NPWT treatment remains under debate. However, granulation tissue growth in the NPWT group was unsatisfactory. The present study compares the effect of regeneration therapy of combining NPWT with PRF to other three methods (normal saline, NPWT and PRF) by testing the bone-exposed area, skin contraction rate, and granulation coverage and the level of growth factors in granulation tissue based on a rat model of bone-exposed wounds. Consistent with previous reports [6], on day 10, only the wounds in the N + P group were completely covered with granulation (Figure 4). Thus, our results suggest that the major contribution of NPWT to wound healing is to promote wound contraction through mechanical force, which we called macrodeformation, rather than stimulate the growth of granulation. Although the PRF shrank to various degrees, PRF clots remained in the wounds of the PRF and N + P groups, indicating that the PRF treatment was available and functional during the entire healing process. Intriguingly, although accelerated granulation tissue growth did not statistically differ between the PRF and N + P groups on day 4, the granulation coverage became significantly greater in the PRF group than in the N + P group on days 7 and 10 (Figure 4). Fibroblasts migrate from surrounding tissue to wounds in response to chemotactic signaling, which is important to the proliferation phase of tissue repair [32]. Cells exposed to NPWT exhibits a higher rates of proliferation and migration [33]. These cells proliferate and apply strong contractile forces across the wound bed, thus drawing the wound edges together [34]. On Masson staining, fibroblasts were also observed in three experimental groups, arranged directionality to a certain extent, with more consistent directionalities in NPWT and N + P groups, indicating that this directional arrangement is the result of the strain force exerted by NPWT evenly distributed on the wound surface. Thus, this reduction in the accelerated growth of granulation tissue in the N + P group could be explained by the increased wound contraction in N + P group, which limited the maximum area of granulation tissue. Taken together, the present results demonstrate that NPWT strengthens the contraction of wounds in the repair of bone-exposed wounds.

TGF-β1 is an important factor in wound healing. It has been demonstrated that, in vitro, NPWT increases TGF-β1 levels in fibroblasts [33,35]. Kilpadi et al. [36] showed that NPTW can upregulate the level of TGF-β1 in acute wounds. In addition, Yang et al. [37] also obtained similar results in the study of NPTW in the treatment of chronic diabetic foot, supporting the view that the mechanism of NPTW in promoting wound healing might be related to the up-regulation of TGF-β1 levels. In the present study, the levels of TGF-β and bFGF in the NPWT group were significantly higher than those in the control group (p < 0.05; Figure 5C & D). Our results show that NPWT facilitated the expression of TGF-β1 in granulation tissue of the bone-exposure wound.

In the present study, the hair in the anterior region of SD rats skull was removed before surgery. However, routine shaving often fails to achieve complete hair removal and is often used in conjunction with other depilation methods. Relevant research shows that the depilatory cream was effective, atraumatic, nontoxic and could be used safely on granulating wounds associated with a significant reduction in skin surface bacteria [16]. Furthermore, depilatory creams have an advantage in areas where shaving is difficult [17]. The combination of routine shaving and depilatory cream was used to maintain the scalp of the rat, a hairless state which achieved good results. And we did it again when we changed wound dressing. Considering the natural wound contraction in rodent species, foamy frames could be fixed firmly to the epidermis to prevent wound shrinking. As mentioned before, a rodent model may exaggerate the contribution of wound contraction. Thus, we speculate that PRF is necessary and more important for treating bone-exposed wounds in clinical practice [38]. It is worth noting that PRF should be applied to the entire wound surface in a thin layer and the NPWT dressing covered it a little wider, otherwise the occlusion of the NPWT caused by the gel would inevitably happen due to the special structure of the PRF.

PRF is a kind of platelet concentration, which resembles PRP. Although PRF cannot achieve the same level of cytokines as that achieved by PRP, the slow polymerization during the preparation process of PRF seems to generate a fibrin network that is very similar to the natural one, which is a clot with a 3D structure under the electron microscopy [39]. This structure can lead to more effective cell migration and proliferation [20]. And the simplicity of its preparation method of not requiring additives such as thrombin is also conducive to its clinical application. On the day 10, we can still detect a higher levels of growth factors in the granulation tissues of the PRF group and the N + P group than those in the control group (p < 0.05; Figure 5). The grid structure attributes to the function of PRF protects growth factors from proteolysis [20]. PRF can consistently and stably release various growth factors [38], such as VEGF, PDGF-BB, TGF-β1 and bFGF. These growth factors are essential for wound healing [40], as VEGF and bFGF are angiogenic factors, while PDGF-BB and TGF-β are profibrotic factors [41].

In 1974, Ross first described PDGF, which exists in platelet α-granule sorgiant cells and stimulates angiogenesis, osteoblastic proliferation and differentiation, as well as mesenchymal cell division. In fibroblast, PDGF also facilitates cell proliferation and collagen synthesis [42]. The growth factors TGF-β and PDGF both are important in the formation of granulation tissue as they can upregulate collagen production, while PDGF also upregulates the synthesis of glycosaminoglycan (GAG) and fibronectin in fibroblasts [7]. In the previous study of negative pressure suction promoting wound healing, an increase in the levels of TGF-β and PDGF [43] and an upregulation of bFGF, TGF-1 expression have been observed [35]. In the present study, we found that the levels of the four growth factors (VEGF, PDGF-BB, TGF-β and bFGF) in the granulation tissue of the N + P group were significantly higher than those of the other three groups (p < 0.01; Figure 5). This result indicates that the combination therapy of NPWT and PRF seems to apply an increased underlying impact of PRF. On day 10, the levels of VEGF and bFGF were significantly higher in the N + P group than those in the control group (Figure 5B & C). Notably, the VEGF level of the N + P group was significantly higher than that of the other three groups, (Figure 5B) indicating that the combination of NPWT and PRF can effectively promote the angiogenesis of granulation tissue. Furthermore, the results of the growth factor analysis were in accordance with the HE staining results. On day 10, there were no significant differences among the groups in the granulation tissue at the edge of the wound area; however, the granulation tissue at the center of the wound area showed significantly more functional vessels in the N + P group than in the other three groups (Figure 4). Additionally, the levels of the profibrotic factors, PDGF-BB and TGF-β, were significantly higher in the N + P group than in the control group. In accordance with the Masson staining results, granulation tissues in the PRF and N + P groups formed much more collagen whose structure was better organized (Figure 4). However, compared with the N + P group, the PRF group is lacking of directional mechanical traction which leaded collagen more even and homogeneous. Base on the results, the production of more granulation tissue should be mainly attributed to the effect of PRF. Taken together, NPWT combined with PRF treatment played a significant role in the late proliferation phase of wound repair and the early tissue remodeling phase.

Most importantly, PRF might provide structured fibronectin for cell adhesion, migration, proliferation and differentiation [12,39,40]. Compared with the N + P group, the PRF group is lacking of directional mechanical traction which leaded collagen more even and homogeneous. Wound repair is based on the proliferation and migration of inflammatory cells, fibroblasts and vascular endothelial cells [44]. Collective cell migration is more efficient than single cell migration in wound repair, and intercellular interactions provide an important biological basis for collective cell migration [13]. The cadherin between cells [45] and the integrin between cells and the extracellular matrix [46] not only maintain the physical integrity of the cell population, but also help coordinate the behavior of adjacent cells [47]. Wang et al. [48] speculated that free cells in the wound margin lose their adhesion with the extracellular matrix, which disturbs the interaction between cells, rather than directly affecting the cytoskeletal dynamics required for cell migration. In the migration process, wound cells form protrusions in random directions, resulting in a significant decrease in the persistence of migration movement [49]. Therefore, when bone-exposed wounds reach a critical size, the contraction of the wound is not enough to heal the wound completely, coupled with the degenerated proliferation and migration of the cells [25], resulting in the wound unable to heal [12,13,50]. With PRF present during the wound-repair process, the leading cells can adhere to the PRF structure [13], thus changing the healing mechanism. This will be confirmed in subsequent experiments.

The mechanotransduction process in which cells transduce mechanical forces into biological signals is also an integral part of the MDWT mechanism [51,52]. The mechanical transduction signaling of connective tissue associated with fibroproliferative diseases has been thoroughly studied [53]. In contrast, the study of mechanical transduction in MDWT is a relatively new area of research. Currently, it is believed to involve molecules such as nitric oxide and other hypoxia pathways [54]. In this study, we observed satisfactory granulation tissue growth, vascularization structure and collagen remodeling in the N + P group. Recently, some studies have shown that the early and continuous activation of mast cells seems to be related to the growth of granulation and the maturation of collagen [31]. However, it is unclear whether mast cells play a role in this combination therapy strategy, and the related links should be further explored in follow-up studies.

Due to a lack of the necessary conditions for cell proliferation and migration, a bone-exposed wound is unlikely to heal naturally. By introducing NPWT and PRF, we achieved satisfactory results in a rat model. The present study demonstrated that NPWT combined with PRF, as a treatment strategy, can accelerate the healing of bone-exposed wounds, shortening the repair time. Furthermore, this treatment strategy can potentially reduce the duration of VSD dressing replacement in clinical practice, and thus, ease the economic and psychological burdens on the patients.

Conclusion

Combination of NPWT and PRF can repair bone-exposed wounds effectively and accelerate wound healing. Our study suggests that the wounds treated by NPWT with PRF increases wound closure rate with significant contraction and granulation. The major contribution of NPWT was wound contraction through macrodeformation, while PRF stimulated granulation growth. This combination strategy also stimulated the release of the PRF-related angiogenic and profibrotic growth factors and increased the centric vessel density of granulation tissue.

Future perspective

Bone-exposed wounds are one of the great challenges that plastic surgeons have to deal with. The lack of periosteum on the bone surface inevitably leads to chronic and refractory wounds. The development of effective bone-exposed wound treatment is of great significance. However, tissue flaps, as a traditionally major therapy used in repairing bone-exposed wounds, is limited by many conditions [22], and inevitably causes defects at the donor site. NPWT has become a widely used treatment for acute and chronic wounds [55]. Generally, the closed and stable environment that NPWT maintained is favorable for granulation tissue growth [7,8]. Although a few studies that utilized NPWT to treat bone-exposed wounds in rodent models have also shown promising results [2,8], the healing in these studies may be mainly attributed to wound contraction, rather than the re-epithelialization or granulation tissue growth [9], which is consistent with clinical observations, as applying NPWT to cover bone-exposed wounds does not seem to work well [2]. NPWT is usually combined with an autografted or allografted dermal substitute [10,11]. PRF contains a variety of cytokines and can provide fibronectin scaffolds [12] directly to lead the migration of inflammatory cells [13,14]. Our study of skull-exposure murine model has confirmed this combination strategy could achieve mesenchymal healing with granulation tissue. It is important to note the molecular mechanisms that occur during bone-exposed wounds repair. Though the histology data could partly explain the differences between edge and center, molecular profiles in wound center and edge should be traced and additional experimentation is required on exploring the possible molecular mechanisms in order to better reveal this regeneration phenomenon. Regenerative therapy of NPWT combined with PRF is a simple method and is expected to be applied clinically for patients with bone-exposed wounds in the future. This new treatment might have a potential impact on reducing the duration of VSD dressing replacement in clinical practice, and thus, reducing the financial and psychological burden on patients.