Role for platelet rich plasma as an adjuvant therapy in wound healing and burns

Platelet rich plasma (PRP) contains high concentrations of growth factors. Intuitively, these were thought to be of potential benefit in healing of chronic wounds, skin grafts and graft donor sites. This was echoed in retrospective studies and an individual case basis but had not been randomized. A systematic search was carried out by two individuals, independently, on the MEDLINE, EMBASE and COCHRANE databases, according to PRISMA guidelines. All data analysis and statistics was pooled and analysed using the Cochrane RevMan Software. Split thickness grafts have been described by numerous authors, but PRP use did not offer a significant advantage in graft take in pooled results (MD 5.83, 95% CI − 0.69 to 12.25, random-effects, p = 0.08). The analysis of included randomized controlled trials has shown favourable split-thickness skin graft donor site healing in the PRP group (MD − 5.55, 95% CI − 7.40 to − 3.69, random-effects, p =  < 0.00001) compared to the control group. For carpal tunnel syndrome, the pooled results showed no difference in functional scores vs steroid injections (SMD − 0.68, 95% CI − 1.47 to 0.10, randomeffects, p = 0.09) or indeed splinting groups. With the current body of evidence, we conclude that the use of PRP as an adjuvant therapy in skin grafts, burns, carpal tunnel surgery or scars cannot be rationalised. A potential use of PRP is in donor site management but the cost of this would be difficult to justify. Level of evidence: Not ratable.


Introduction
Platelet rich plasma was initially studied in the 1970s in the field of haematology for its use in blood transfusions for patients with thrombocytopenia. It is an autologous treatment for soft tissue healing and bone healing which has been present for decades [1] and has been used to treat wounds since 1985 [2]. Since then, it has grown in popularity in many surgical specialties including plastic and reconstructive surgery.
Platelet rich plasma (PRP) is defined as an autologous blood-derived fraction containing high concentrations of platelets and growth factors [3]. The growth factors identified in PRP include a platelet-derived growth factor (PDGF), transforming growth factor β (TGF-β), vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), insulin-like growth factor (IGF) and fibroblast growth factor (FGF) [ Table 1]. The red blood cells are removed through a process of centrifugation, leaving behind a high concentrate 1 3 of platelets, fibrinogen and growth factors which include TGF, EGF, PDGF and VEGF [3].
Other autologous blood-derived products have been developed as a result of the popularity of PRP. Platelet-rich fibrin (PRF) comprises the second generation of platelet concentrates, with its high leukocyte concentration improving tissue healing and recovery [4]. It is preferred by some over conventional PRP due to not requiring bovine thrombin or anti-coagulants in its formation, thus not interfering with the normal clotting cascade [4,5]. Human platelet lysate represents another variation of blood derivatives used in regenerative medicine. In particular, its high concentrations of neurotrophin 3, nerve growth factor, and brain-derived growth factor have made it an attractive additional option [6].
Even within PRP itself, there has been some variation. Leukocyte rich PRP (LR-PRP) and leukocyte poor PRP (LP-PRP) have both been used, with elevated leukocyte count being associated with increased pro-inflammatory effects. This is in part due to elevated catabolic cytokines such as interleukin 1 β, tumour necrosis factor α and metalloproteinases [7]. PRP remains the most researched blood product derivative to date and will form the centrepoint for this review.
In addition to its application in various specialties including aesthetics, orthopaedics, oral and maxillofacial, vascular and cardiac surgery, increasing interest is seen in the application of PRP in tissue regeneration, such as soft-tissue defects and skin grafting [8,9], as well as acute and chronic wound management [10].
Even so, the literature to date has failed to provide definitive evidence on whether there is a significant role of PRP as an adjuvant therapy in plastic surgery-related wounds.
This systematic review and meta-analysis will attempt to summarize randomized controlled trials (RCTs) up to 2020 comparing PRP and conventional treatments in plastic surgical wound management.
Narrative syntheses [11] of literature have looked favourably on the use of PRP as an adjunct in the treatment of patients in plastic and reconstructive surgery. Retrospective studies and case series have shown favourable results for adjuvant PRP use but the same findings are not as prevalent in randomized controlled trials (RCTs).

Literature summary
Numerous randomized controlled trials, reviews and metaanalyses on platelet rich plasma as an adjuvant therapy have been published with various findings [11][12][13]. We focus our study to its uses in plastic and reconstructive surgery, namely in common areas of hospital-based practice such as wound healing, skin grafts, carpal tunnel syndrome and burns.

Split-thickness skin grafts and donor site healing
Narrative reviews have cited the benefits of adhesive properties in PRP for split-thickness graft adherence and take [11]. Surprisingly, two of five included studies in this review did not report on graft take (the primary outcome) and were deemed a high risk of bias. The reported rates of split-thickness graft take in the control groups were unaccountably lower (15-76%) [11] than literature reported/ expected rates of take (90-95%). This begs the question Platelet derived growth factor PDGF Stimulation of collagen synthesis, proliferation of bone cells, fibroblast chemotaxis and proliferative activity, macrophage activation Transforming growth factor ß TGF-ß Enhances synthesis of type I collagen, promotes angiogenesis, stimulates chemotaxis of immune cells, inhibits osteoclast formation and bone resorption Vascular endothelial growth factor VEGF Stimulates angiogenesis, migration and mitosis of endothelial cells, increases permeability of the vessels, stimulates chemotaxis of macrophages and neutrophils Epidermal growth factor EGF Stimulates cellular proliferation, differentiation of epithelial cells, promotes cytokine secretion by mesenchymal and epithelial cells Insulin-like growth factor IGF Promotes cell growth, differentiation, recruitment in bone, blood vessel, skin and other tissues, stimulates collagen synthesis together with PDGF Fibroblast growth factor FGF Promotes proliferation of mesenchymal cells, chondrocytes and osteoblasts, stimulates the growth and differentiation of chondrocytes and osteoblasts whether the selection of patients/wounds for split-thickness grafting was appropriate in the first instance and whether the study methodology was appropriate.
There have been studies which looked at the use of PRP as an adjunct to accelerate the healing of split-thickness graft donor sites. A pooled analysis of five RCTs [14] has shown that the healing of the donor site was faster than in the control group and that the pain score (measured on the VAS) at the time of dressing change was lower in the PRP group. The difference in discomfort has been attributed to the PRP gel applied in the treatment group, avoiding adherence of the traditional dressings. The results showed significant heterogeneity, however, likely due to differences in outcome reporting, variation in PRP preparations and study design.

Burns
Similar to carpal tunnels, chronic wounds and graft take, the utility for PRP in burns is questionable [15]. A metaanalysis found evidence which shows improved healing time and epithelialization in the PRP group when compared with the control. However, the findings were subject to very high levels of clinical and statistical heterogeneity (I 2 > 90%). In addition, the highest quality RCT in the pooled results (DB-RCT) was inexplicably excluded from the pooled analysis which raises concerns with the validity of the findings. There was no meta-regression or sensitivity analysis performed to explore the nature of these differing results.

Carpal tunnel
The use of PRP injections in place of corticosteroids for the treatment of carpal tunnel syndrome is a novel hypothesis. A meta-analysis of four controlled trials [16] has shown improved symptoms and pain in the PRP cohort when compared to the control group, but no difference in the functional outcome. Of note, 3 of the RCTs (75%) used either splinting or saline as a control group rather than the proven treatment, corticosteroids. This would exaggerate the benefit of PRP as the comparator is not a recognised medical intervention. In addition to this, moderate to severe cases were excluded from the treatment arms, which may further exaggerate the efficacy of PRP as a treatment modality. The included studies were identified as carrying a significant risk of bias. A larger, more comprehensive meta-analysis [17] of 9 RCTs showed no difference in the VAS or Boston carpal tunnel questionnaire (BCTQ) at 1-month follow up. Significant differences were reported in electrophysiological differences, but no significant difference in pain or function was identified.

Wound healing
PRP has been utilized in the management of diabetic ulcers since the 1990s. A meta-analysis of eight randomized controlled trials [13] on its use in the treatment of diabetic ulcers has shown an increased likelihood of complete ulcer healing with adjuvant PRP, but no difference in the recurrence rate or amputation rate. In the clinical management of sternal wound dehiscence adjuvant PRP was shown to reduce further incidence of sternal wound dehiscence. A meta-analysis [12] has shown a favourable response in the retrospective studies subgroup; however, the clinical benefits were not observed in the randomized controlled trial.

Study aims and objectives
The primary aim is to identify whether there's a role for platelet rich plasma as an adjuvant therapy in plastic and reconstruction surgery, namely in wound healing, grafting, mechanical neuropathies, keloid/hypertrophic scars or burns.
To determine this, we will consider applicable key outcomes from the aforementioned topics. The study will follow the PICOs framework.

Methods
The methodology follows the PRISMA guidelines, and our protocol is accessible in the PROSPERO database https:// www. crd. york. ac. uk/ prosp ero/. A systematic search was carried out by two individuals, independently, on the MED-LINE, EMBASE and COCHRANE databases [ Fig. 1].
Abstracts were screened for eligibility, and full text papers were reviewed to ensure they meet the inclusion criteria. The summary of the search results is illustrated on the PRISMA flow sheet (Fig. 1). All studies were critically appraised using the systematic GRADE criteria.
Inclusion criteria: • Randomized controlled studies • Prospective cohort studies • Platelet rich plasma therapy used in plastic surgery as an adjuvant therapy to accelerate or precipitate a favourable outcome, namely in wound healing, burns, skin grafts and mechanical neuropathies.
Exclusion criteria: • Studies which were no relevant to plastic and reconstructive surgery • Studies which failed to report outcome results in detail for inclusion in analysis • Case reports/case series 1 3

Data sources
The data was extracted by the primary author and secondary author and verified by a senior specialist to ensure integrity. Any disagreements and differences will be resolved by discussion or referral to a third reviewer.
Data from the retrieved studies were tabulated according to: (i) Study characteristics (ii) Cohort characteristics (iii) Method of preparation: PRP harvesting techniques, volume applied and number of applications (iv) Outcome of interest (v) Results (vi) Adverse events

Data Analysis
All data analysis and statistics were pooled and analysed using the Cochrane RevMan software. Results were reported as mean or standardised mean differences with confidence intervals. Overall effect was determined with the p-value. A result of p = < 0.05 was considered significant. The test for outcome variance (I 2 ) was included in the results where it exceeded the 30% threshold.

Excluded studies
In evaluating PRP in the management of carpal tunnel syndrome (CTS), two of the nine identified studies used placebos as their control group. One did not report follow up results and could not be included.

Split-thickness skin grafts
Multiple authors reported their results on PRP use in splitthickness skin grafts. Four included studies [18][19][20][21] reported their skin graft take results. Grafts were either designated as successful or unsuccessful on their clinical review (Figs. 2, 3, 4, and 5). They reported their success rate as a percentage. The pooled results (n = 98) showed no significant findings in the overall percentage of skin graft success between the two comparison groups (MD 5.83, 95% CI − 0.69 to 12.25, random-effects, p = 0.08) (Fig. 6). Subgroup analysis of PRP and graft take in burns also failed to show any meaningful results (MD 1.04, 95% CI − 6.19 to 8.27, random-effects, p = 0.78; I 2 = 0%) (Fig. 7). Only two studies [19,22] reported healing time in days. There was no statistically significant difference in graft healing time in the two groups (MD − 17.22, 95% CI − 48.12 to 13.68, random-effects, p = 0.27) (Fig. 8). There was significant heterogeneity in these findings (I 2 = 74%).
Epithelialization rates were also reported in three studies. Results were pooled and reported with a standardized mean difference as one study [23] reported epithelialization in millimetres (mm) and the remaining two [20,24] reported epithelialization as a percentage from the overall wound site. These reported results were all from the 1-week interval. No differences were found between the groups (SMD 0.29, 95% CI − 0.40 to 0.98, random-effects, p = 0.41) (Fig. 3). Similarly, heterogeneity was high in these results (I 2 = 77%).

Discussion
The use of PRP in improving split-thickness skin graft take has failed to show any promise. We do highlight that PRP may confer a benefit in accelerating split thickness donor site healing (p = < 0.00001). Whether this is practical is questionable as the overall morbidity from the donor site is low, and the health economics argument would favour the more affordable treatment as the final outcomes is ultimately the same and does not influence length of inpatient stay.
There were studies measuring the effect of PRP on burn healing in conservative management and in skin grafts [37][38][39][40]. However, we did not pool these results. The reporting of outcomes in the identified studies [37,39] was incomplete, limited and too varied in order to pool the data coherently. Two meta-analyses [15,41] summarising the effects of PRP on burns were published in 2020, favouring the use of PRP to accelerate burn healing. However, we could include these studies or results due to significant concerns about the nature of the included studies and in these meta-analyses. Some included articles did not have an official translation, so verifying the results was not possible. Others compared the use of an acellular dermal matrix (ADM) with and without PRP [42]; therefore, pooling these results with studies comparing conservative healing with and without PRP would be invalid. Furthermore, studies which contributed favourably towards PRP use were paediatric cohorts [43]. This explains the exceedingly high heterogeneity in the published results (I 2 = > 85%), casting doubt to whether the significant findings are valid or reproducible. Arguably, the highest quality study (DB-RCT) [20] identified in the published meta-analysis [15] was inexplicably excluded. Overall, we found no meaningful or forthright way in which to pool these results while adhering to methodological principles. Studies which reported a benefit were fraught with inconsistencies, poor reporting of key outcomes and questionable study methodology. High-quality evidence [20] failed to show any benefit.
We found no supporting evidence for PRP use in carpal tunnel syndrome. Symptom severity and function were our key outcomes. Due to limited reporting and exclusion of one study [44], we failed to achieve results for the "PRP vs placebo" subgroup when investigating PRP use in CTS. The individual studies did not report any difference between the groups in their results. Previously published meta-analyses [16,17] have identified results supporting PRP use in CTS, but on closer examination of the summary of data and results, we identified that all control cohorts (splints, corticosteroids and placebos) were grouped together as one which is methodologically erroneous, concluding that these findings are imprecise. When these subgroups are pooled more appropriately, "PRP vs Placebo", "PRP vs Splint" and '"PRP vs Corticosteroid", we found no significant findings. In addition to this, there were subtle methodological errors; pooling data with different outcome measures with "mean difference" instead of "standardised mean difference", and using "fixed effects" models instead of "random effects", which is the correct metric when combing different populationbased data.
We could not perform a quantitative analysis on PRP use in acute or chronic wounds. Wound healing is a complex process which is highly dependent on both patient and environmental factors. Although some results on wound healing have been published in the literature, the variance in wound types, cause, locations, environment and patient factors contribute to the great deal of confounding factors in these studies. Any pooled data would carry significant clinical and statistical heterogeneity and would call any results into question. Furthermore, there is a significant lack of objective outcome reporting in these studies which would make meaningful analysis more difficult.
We do note there are recent studies [45][46][47][48][49] which demonstrate some evidence for blood derivatives (PRP) being used in wound healing and severe burns. These studies did not fit our initial inclusion/search criteria but their results could provide early suggestion for some clinical benefit.

Limitations
Our results and findings are contingent on reported results in included studies. The included studies often demonstrated differing results for the same outcomes. The data analysis for outcomes of interest frequently had high heterogeneity. Inclusion of RCTs as well as prospective clinical studies slightly undermines the otherwise higher quality of evidence with meta-analyses of exclusively RCTs.

Conclusions
It is hard to be certain of whether there is any benefit to PRP use as an adjuvant therapy in plastic surgery. The current quality of evidence supporting its use in limited circumstances is low. High-quality studies showed no difference in its application when compared to control groups. A high proportion of studies supporting its use comes from low level evidence with questionable study design and outcome reporting. There have been a few recent articles which show some potential and benefit in some specific clinical applications in the domain of wound healing in burns. With the current body of evidence however, we cannot support its use as an adjuvant therapy. There might be some role in using PRP for donor site healing but with low morbidity and difference in cost of treatment, and it would be hard to make an argument for its use based on health economics.

Declarations
Ethics approval and consent to participate Ethical approval and consent are not required for this review article. This article does not contain any studies involving human participants or animals performed by any of the authors. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.