Abstract
Objective:
Autologous skin transplantation is limited by donor site availability for patients with extensive burns. The objective of this study was to demonstrate the feasibility and efficacy of split-thickness skin (STS) and dermal pixel grafts (PG) in the treatment of burns.
Approach:
The study was divided into three arms of validation, expansion, and combination that all followed the same study design. Sixteen deep partial-thickness burns were created on the dorsum of anesthetized pigs. Three days postinjury the burns were debrided and grafted with STS and dermal PGs. The PGs were prepared by harvesting two skin grafts (split-thickness skin graft [STSG] and dermal graft) from the same donor site going down in depth. The grafts were minced to 0.3 × 0.3 × 0.3 mm PGs and suspended in a small volume of hydrogel. Healing was monitored for 6, 10, 14, or 28 days. In the validation study the PGs at 1:2 expansion ratio were transplanted and compared with STSG and untreated controls. The expansion study investigated the maximum expansion potential of the PGs and the combination of the benefits of transplanting STS and dermal PGs together.
Results:
The validation study showed that when STS and dermal PGs were transplanted in a 1:2 ratio they fully re-epithelialized the wounds in 14 days. The expansion study demonstrated that using expansion ratios up to 1:500 the wounds were re-epithelialized by day 28. The combination study showed that there was no additional benefit to use STS and dermal PGs together.
Innovation:
Pixel grafting provides expansion ratios greater than conventional STSG. The possibility to harvest both STS and dermal PGs from the same donor area further reduces the need for healthy skin.
Conclusion:
STSG and dermal grafts can be minced to PGs with preserved viability and expanded up to 500 times to re-epithelialize a wound.
INTRODUCTION
Large burn injuries present a complex physiological and surgical challenge. After initial medical stabilization of the patient with a combined anesthesiology and burns physician approach; the next step is to consider treating the burn itself and getting the injured area healed. Burn injuries account for >9 million causalities globally per annum with ∼0.5 million in the United States alone. 1,2
Jackson described the zones of the burn, with the zone of coagulation being the worst affected area with tissue loss/damage being irreversible. 3 Thus, deep and large burns often require hospitalization and surgical treatment. The basic principles of the surgical management of burns are to debride necrotic or nonviable burnt tissue as it is an infection risk and delays the healing of the wound. 4
Autologous skin grafting is the standard of care (SoC) approach for covering the wound to ensure closure and good quality of healing. In particular, split thickness skin grafting (STSG) which involves taking the epidermis and a part of the dermis as a graft from healthy donor tissue and transplanting that as an autograft to the burn wound. 5 A limitation in large burn injuries is finding enough healthy donor skin to graft onto the burn. In addition, there is morbidity associated with donor sites such as pain, infection risk, and discoloration. Current SoC involves meshing the STSG, which can allow expansion the graft up to six times (possibly nine times) its original size. 6
However, large expansion of the STSG often results in poor healing outcomes. 7 Several alternative techniques to expand donor skin such as cultured epithelial autografts (CEA), full-thickness skin column transplantation, and autologous cell sprays have been developed but none are optimal. 4,8 –10
Pixel grafting is a novel technique that maximizes the use of donor skin, whereas minimizing donor site morbidity. 11 The methodology relies on mincing a small area of donor skin using a handheld mincer to create tiny skin pieces, each of which will serve as a small center of skin regeneration. 12 Importantly, it has been shown both in preclinical and clinical studies that in a moist wound environment, the orientation of the graft within the wound bed does not affect the overall healing outcome making pixel grafting more practical than for example, Meek's technique where the micrografts must be placed with the dermis side down. 11,13
Furthermore, pixel grafting can be done under local anesthesia in a procedure room, an examination room, or even at the bedside. Our previous studies have shown that graft take appears to be the same as with meshed STSG. 11,13,14
Skin adnexal structures such as hair follicles and sebaceous glands are located in the dermis. These skin appendages have been shown to contain epidermal stem cells (ESC) found in the hair follicles and sebaceous glands. 15 ESCs differentiate based upon signals they receive, and during the migratory phase of wound healing the ESCs differentiate into epidermal cells playing a vital role in the healing process. 16,17
Thus, it has been shown that tissue from the dermis can be used in skin grafting with similar healing and re-epithelialization results to STSG, including the epidermis. 18 –21 Several clinical studies, utilizing dermal grafts in the treatment of burns, have reported good graft take and wound closure. Also, no difference in donor site healing has been observed when compared with STSG. 22,23 We have previously demonstrated that micrografts derived from dermis contain ESCs and grafting with them resulted in a healed wound comparable with grafts derived from epidermis with a portion of dermis. 24
The objective of this preclinical project was to demonstrate the feasibility and efficacy of STS and dermal PGs in the treatment of full-thickness burns. The study was divided into three arms of validation, expansion, and combination. The purpose of the validation study was to evaluate the feasibility of STS and dermal PG transplants mixed in a hydrogel and compared with 1:2 expansion of these grafts to the current SoC (STSG) and untreated control.
The purpose of the expansion arm of the study was to investigate the optimal and maximal expansion ratios for both STS and dermal PGs and compare them with each other in terms of wound healing and quality of healing. The purpose of the combination study was to explore if transplantation of 1:1 mix of STS and dermal PGs would have any benefits especially in terms of quality of healing.
INNOVATION
Pixel grafting can provide expansion ratios much greater than conventional STSG. The possibility to harvest both STSG and dermal pixel grafts (PGs) from the same donor area further reduces the need for donor site surface area. The grafts can be placed randomly with full graft survival as long as they are allowed to heal in a moist environment. The quality of the healed wound appears to be as good as that of a meshed STSG. In the outpatient setting pixel grafting can be done under local anesthesia in a procedure room, an examination room, or even at the bedside. 11
CLINICAL PROBLEM ADDRESSED
Large burns from explosions, fire, and warfare present a global challenge for physicians to manage safely. 25 Donor site availability and maximum expansion ratios of STSGs are limiting factors in the skin grafting of major full thickness burns. 5 Various alternative techniques that could enable larger expansion of STSGs and minimize donor site morbidity have been introduced over the years. However, they are often accompanied with high costs and legislative challenges while still being limited in their clinical efficacy. The objective of this study was to investigate the feasibility and efficacy of both STSG and dermal PGs in the treatment of full-thickness burn wounds. 26
MATERIALS AND METHODS
Animals, anesthesia, and analgesia
The animal experiments were performed under good laboratory practice at Toxikon Corporation (Bedford, MA) and were approved by the Animal Care and Use Committee (IACUC, #2019-NR-34) and the Animal Care and Use Review Office (ACURO # BA170661.e003). In total 16 female Yorkshire pigs, weighing at least 50 kg were used. On the day of surgery, the animals were anesthetized using ketamine ∼33 mg/kg, acepromazine ∼1.1 mg/kg, xylazine ∼2.2 mg/kg, and atropine ∼0.02 mg/kg. Anesthesia was maintained with isoflurane 0–5% through face mask or endotracheal tube. Buprenorphine was given presurgery at 0.03 mg/mL. Fentanyl patches (sufficient to deliver 2–3 mcg/kg/h) were placed for pain control.
Porcine model
Day 3
The burn wounds were created using a custom burn device (Harvard School of Engineering, Cambridge, MA) to create standardized full-thickness burns. 27 Each pig received, on its dorsum, 16 full-thickness burn wounds, each measuring 2.5 cm diameter. Burn edges were outlined using a tattoo gun (Spaulding & Rogers Mfg., Inc., Voorheesville, NY) to allow evaluation of wound contraction. At the end of the procedures, burns were covered with Tegaderm film (3M; Maplewood, MN) (Fig. 1).

Study design. Full-thickness burns were created on the dorsum of pigs. Three days later the burns were surgically debrided and grafted with split-thickness skin and dermal pixel grafts. The animals were followed up to either 6, 10, 14, or 28 days and both wound healing and quality were evaluated. This same study design was used in the all arms of validation (4 animals), expansion (8 animals), and combination (4 animals) of the study.
Day 0
The burn wounds were surgically debrided by removing eschar and necrotic tissue using a scalpel. An STSG (0.3 mm thick) and a dermal graft (0.3 mm thick) were harvested in layers from the same donor site in the buttock region of each pig with a pneumatic dermatome (Zimmer Biomet, Warsaw, IN). The donor site was covered with a Tegaderm film dressing (3M).
The STSG and the dermal graft were minced to 0.3 × 0.3 × 0.3 mm PGs using an Xpansion ® micro-autografting kit (Applied Tissue Technologies, Hingham, MA). Subsequently, STS and dermal PGs were mixed in hydrogel (Solosite R; Smith & Nephew, London, United Kingdom), transplanted in a small volume (500 μL) at different expansion ratios (Table 1) evenly over the base of each wound. All wounds were covered with Platform Wound Devices (Applied Tissue Technologies) (Fig. 1).
This study was divided into three arms of validation, expansion, and combination
In each arm 16 burns were created on each pig. This table shows the treatment groups, time points, and n-numbers for each arm.
STS, split-thickness skin; STSG, split-thickness skin graft; d, day; SoC, standard of care.
Days 6, 10, 14, and 28
The wounds were photographed (Silhouette Star™; Aranz Medical Ltd., New Zealand) and subsequently euthanized by an injectable barbiturate (Euthasol; Virbac Corporation, Westlake, TX) either on day 6, 10, 14, or 28 post-treatment. A full-thickness excisional biopsy (wound strip) was harvested from each wound immediately after euthanasia to give a cross-sectional view of the wound edge-to-edge (Fig. 1).
Treatments
This project was divided into Validation, Expansion, and Combination studies. In the Validation study (4 animals, 16 wounds/animal), wounds were randomized to receive either STSG, STS PGs (1:2), dermal PGs (1:2), or gauze. In randomization, the experimental block design was used to increase the power and reproducibility, as well as to minimize variations and experimental errors. 28 N-number was 4 per treatment group at each time point (Table 1).
In the Expansion study (8 animals, 16 wounds/animal), wounds were randomized and treated with various expansion ratios (1:10, 1:50, 1:100, or 1:500) of either STS PGs or dermal PGs. N-number was 4 per treatment group at each time point (Table 1).
In the Combination study (4 animals, 16 wounds/animal), wounds were randomized and treated with 1:1 combination of STS PGs and dermal PGs at 1:10, 1:100, and 1:500 expansion ratios. N-number was 4 per treatment group at each time point (Table 1).
Histology
The harvested wound strips were fixed in 4% neutral buffered formalin, embedded in paraffin, and sectioned for staining with hematoxylin and eosin (H&E) and Masson's trichrome.
Analyses
Re-epithelialization on days 6, 10, 14, and 28 post-transplantation was quantified as the area of the new epithelium divided by the wound area. Epidermal thickness on day 28 post-transplantation was measured in five representative areas of the newly formed epidermis for each wound cross section. The number of rete ridges per millimeter of epidermis was counted from five standardized locations in each wound on day 28 post-transplantation.
Wound contraction was measured of the tattooed margins from macroscopic wound photos using Image J software (NIH). The area inside the tattooed line on was measured and expressed as a percentage of its original size on day 3. All the analyses were performed in a blinded manner.
Statistical analysis
Statistical analysis was conducted using GraphPad Prism 9.3.1 (Graph Pad software, Inc., La Jolla, CA). Data are presented as mean ± standard error of mean (SEM). Comparison of treatments was performed using unpaired one-way ANOVA with Tukey's multiple comparison test. p-Values <0.05 were considered statistically significant. N number for each treatment group was 4.
RESULTS
Validation study
Wound healing
Histology demonstrated that when both STS PGs and dermal PGs were transplanted to the porcine full-thickness wounds in a 1:2 ratio they were able to fully (100%) re-epithelialize the wounds in 14 days. No statistically significant differences between the PG groups (STS PGs and dermal PGs) and the SoC groups (STSG) group were observed. However, in comparison with the untreated wounds all the transplantation groups were statistically significantly more re-epithelialized by day 14 (Fig. 2A, B). Representative macroscopic images of the wounds are shown in the Supplementary Fig. S1.

Validation study. The purpose of the validation study was to evaluate the feasibility of STS and dermal PG transplants mixed in a hydrogel and compared with 1:2 expansion of these grafts to the current standard of care (STSG) and untreated control. In total, 16 burns were created on the dorsum of 4 pigs. On day 3 postburn, the burns were debrided and grafted with STS PG 1:2, dermal PG 1:2, STSG or left untreated (n-number = 4/treatment/time point).
Quality of healing
All the transplanted wounds contracted statistically significantly less than the untreated wounds. No statistically significant differences were observed between STS PG, dermal PG, and STSG groups (Fig. 2C). The results also demonstrated that transplanted wounds had more mature epidermis at day 28 post-transplantation. All the transplantation groups had more rete ridges and thicker epidermis in comparison with the untreated wounds. No statistically significant differences between the treatment groups were observed except that the STS PG-treated wounds demonstrated significantly thicker epidermis than the dermal PG-treated wounds (Fig. 2C, D).
Expansion study
Wound healing
When using expansion ratios from 1:10 to 1:500, no statistically significant differences were observed between the PG study groups in any of the time points. The results showed that by day 14 the re-epithelialization percentage varied in STS PGs-treated wounds from 48% (1:500) to 64% (1:10) and in dermal PGs-treated wounds from 46% (1:10) to 59% (1:100). By day 28 all the wounds were fully re-epithelialized.
Also, no statistically significant differences between the STS and the dermal PGs were observed. In comparison with the SoC, the results showed that by day 14 there were no statistically significant differences between the STS and dermal PG study groups and the STSG (Fig. 3A, B). Representative macroscopic images of the wounds are shown in the Supplementary Fig. S1.

Expansion study. The purpose of the expansion arm of the study was to investigate the optimal and maximal expansion ratios for both STS and dermal PGs and compare them with each other in terms of wound healing and quality of healing. In total, 16 burns were created on the dorsum of 8 pigs. On day 3 postburn, the burns were debrided and grafted with different expansion ratios of STS PGs (1:10, 1:50, 1:100, and 1:500) and dermal PGs (1:10, 1:50, 1:100, and 1:500; n-number = 4/treatment/time point).
Quality of healing
The results showed that on day 28 wound contraction percentage in the STS PG-treated groups varied from 25% (1:50) to 49% (1:100) and from 28% (1:50) to 49 (1:100) in the dermal PG-treated wounds. No statistically significant differences were observed in wound contraction percentage between the different expansion ratios nor the STS and dermal PG study groups.
When compared with controls, the results demonstrated that the STS PG 1:50 and dermal PG 1:50 contracted statistically significantly (p < 0.05) less than the untreated wounds. Thickness of the newly formed epidermis varied from 180 μm (1:100) to 219 μm (1:10) in the STS PG-treated groups and from 178 μm (1:50) to 249 μm (1:500) in the dermal PG-treated wounds. No statistically significant differences between the treatment groups were observed.
In comparison with the SoC control (STSG), wounds transplanted with dermal PGs 1:50 and STS PGs 1:100 had statistically significantly thinner epidermis on day 28 post-transplantation. Also, no statistically significant differences in the number of rete ridges between the treatment groups were observed. The number of rete ridges/mm varied from 0.6 (1:500) to 1.4 (1:50) in the STS PGs-treated wounds and from 0.5 (1:100) to 2.5 (1:50) in the dermal PGs-treated wounds. When compared with the untreated wounds, the wounds grafted with dermal PGs 1:50 demonstrated statistically significantly more rete ridges/mm (Fig. 3C, D).
Combination study
Wound healing
By day 28 all the wounds were fully re-epithelialized. On day 10 the STS + dermal PG 1:10 and STS + dermal PG 1:500 had re-epithelialized statistically significantly more than the untreated wounds. On day 14, the difference between the STS + dermal PG 1:10 and the untreated and the STS + dermal PG 1:100 and the untreated was statistically significant. No statistically significant differences between the PG groups were observed (Fig. 4A, B). Representative macroscopic images of the wounds are shown in the Supplementary Fig. S1.

Combination study. The purpose of the combination study was to explore if transplantation of a mix of STS and dermal PGs would have any benefits for wound healing or quality of healing. In total, 16 burns were created on the dorsum of 4 pigs. On day 3 postburn, the burns were debrided and grafted with 1:10, 1:100, or 1:500 expansion ratios of 1:1 mix of STS PGs and dermal PGs or left untreated (n-number = 4/treatment/time point).
Quality of healing
Regarding wound contraction percentage the difference between the STS PG + dermal PG 1:10 and the untreated was statistically significant. No significant differences were observed between the PG study groups. No statistically significant differences in epidermal thickness or number rete ridges between the groups were observed (Fig. 4C, D).
Donor site healing
The donor sites from where both STSs and dermal grafts were harvested, healed well with minimal scarring and contraction. They were fully re-epithelialized by day 10 after skin graft harvesting.
Discussion
Autologous STS grafting is still regarded as the gold standard for establishing healing of a full-thickness skin wound. In cases of smaller burns when healthy donor skin is available, it provides efficient and aesthetically acceptable wound closure. However, in cases of large burns when donor skin is limited, largely meshed STSGs are associated with substantial problems, such as poor graft take, infection, slow healing and poor aesthetic outcomes. 29
Current emerging alternative techniques to STSG include cell therapies and different biomaterials. ReCell® is a method of delivering autologous epidermal cells from an STSG in the form of a spray. 30 It has been reported as useful in treating partial thickness wounds and is currently not indicated as a single treatment for full thickness wounds. 31 Other autologous cell therapies include CEA. The CEAs are prepared by isolating autologous keratinocytes from a small piece of skin, expanded in vitro and ultimately transplanted back to the patient as sheets. 32
This technique is very laborious and time consuming. The CEAs are also very fragile and sheet take to the wound bed might be problematic. 33,34 Bioengineered skin substitutes are another emerging treatment option that can be used in the management of large burns. They can be acellular such as Integra® or cellular such as Dermagraft®. 35 –37 These products are dermal substitutes and, therefore, a limitation is the requirement for transplantation of an autologous epidermal component most commonly donor skin, which may be limited in large burns. Another limitation is the high cost. 38,39
This study investigated the feasibility to utilize tiny autologous PGs in the management of surgically debrided full-thickness burns. The PGs were derived from both STSG and dermis grafts to make STS and dermal PGs respectively. Subsequently, they were mixed in a small volume of hydrogel to facilitate implantation. Our results demonstrated that using low expansion ratios both STS and dermal PGs were able to fully re-epithelialize the debrided full-thickness porcine burn wounds in 14 days with no statistical difference to SoC treatment.
Similarly, Singh et al. have previously shown that full-thickness wounds grafted with 1:2 STS PGs grafts were able to close the wounds in 10 days. 11 When larger expansion ratios (up to 1:500) were used, the wounds were fully re-epithelialized by day 28. This study and our previous studies have shown that once transplanted in the wound bed each PG becomes its own center of regeneration and the cells start migrating from the grafts to close the wound. Therefore, the wounds that were grafted with high expansion ratios have to rely more on cells from the wound edges that may result in slower re-epithelialization. 11,24,40
In contrast, Hackl et al. transplanted STS micrografts (0.8 × 0.8 × 0.3 mm) to full-thickness wounds with a 1:100 expansion ratio and demonstrated 100% re-epithelialization already on 14 days. 40 In this study, the STS pixel grafted wounds with a 100-fold expansion exhibited ∼60% re-epithelialization on day 14. This discrepancy might be related to the differences between the healing environments of excisional wounds and burns, the burn wound bed being more inflamed and challenging environment for healing. 41,42
In terms of quality of healing no significant differences were observed between the PG groups and SoC. However, our results demonstrated a thicker epidermis and more epidermis to dermis anchoring rete ridges in the wounds treated with the STS PGs compared with dermal PGs-treated wounds. This may be due to the increased number of epidermal keratinocytes in the STS PG groups. The thicker more mature epidermis and presence of more rete ridges seen across all transplanted groups compared with nontransplanted groups is similar to what was seen in a study by Singh et al. 11
However, other similar studies by Nuutila et al. and Hackl et al. that utilized micrografts showed no significant difference between treatment and control groups in terms of maturation of the newly formed epidermis. 24,40 In the combination study, we trialed a combination of STS PGs and dermal PGs (1:1) at various expansion ratios.
The results suggested that there is no additional benefit of using a mixture STS and dermal PGs, since transplantation of either STS or dermal PGs alone results in similar outcomes. However, to really see the effect of these treatments on scarring, the animals should have been followed for >28 days. Another factor is the small burn size that does not replicate the clinical challenge with the larger wounds.
Interestingly, our results demonstrated that dermis grafts can be minced and implanted to achieve wound closure in a similar manner as STS PGs and STSG. We have previously investigated the possibility to use micrografts derived from dermis to close a full-thickness wound. The purpose of that study was to locate the ESCs in the dermis by harvesting sequentially deeper dermal grafts and explore their potential to re-epithelialize a full-thickness wound. Four sequential 0.35 mm-thick skin grafts were harvested from the same donor site going down to 1.4 mm in depth (Layers 1–4).
Subsequently the dermal grafts were minced to 0.8 × 0.8 × 0.3 mm micrografts, transplanted onto full-thickness porcine wounds and healing was compared with wound grafted with STS micrografts. The results concluded that the dermis grafts harvested at the depth of 0.35 mm contained the most CD34+ ESCs and when the wounds were grafted with these dermis grafts no differences to the wounds grafted with STS micrografts were observed. Regarding quality of healing also no significant differences between the groups were observed, such as in this study. 24
In summary, pixel grafting can provide much higher expansion ratios than conventional grafting techniques. From a surgical point of view, pixel grafting is relatively straightforward and does not utilize costly equipment such as some of the emerging advanced wound care therapies. In addition to this, it benefits from not inserting a foreign body into the wound site (such as artificial skin substitutes).
The treatment is fast to deliver because the grafts can be placed randomly–dermis up or down–with full graft survival as long as they are allowed to heal in a moist environment. Pixel grafting can be done in an outpatient setting, which is advantageous compared with conventional split thickness autografting. It has the potential to become a “game changer” in the care of large burns.
Study limitations
In this study, the pigs were euthanized at each time point allowing harvesting of the whole wound to give a cross-sectional view of the wound edge-to-edge for histological analysis instead of collecting tiny biopsies over time. This design was chosen since the tiny biopsies do not give a good representation of the wound and might give unreliable data of the healing rate. In addition, harvesting of the biopsies interferes the healing process. The downside of this design is that it limits the number of biological replicates.
Therefore, to increase the power and reproducibility, as well as to minimize variations and experimental errors, the randomized block design was used to allow comparison of wounds receiving various treatments within the same animal. 28 Furthermore, our previous experiments using this same model have shown that acute wounds in young and healthy pigs heal in a very similar manner. 24,25,43
KEY FINDINGS
PGs can be expanded up to 500, whereas normally STSG cannot be expanded more than 1:6.
PGs provide similar healing rates as SoC (STSG).
No significant differences in STS and dermal PGs in terms of wound healing or quality of healing were observed.
Donor sites heal well although two grafts are being harvested from the same site.
In comparison with cell therapies, this technology is fast and does not have any regulatory issues.
Footnotes
ACKNOWLEDGMENTS AND FUNDING SOURCES
This material is based upon work supported by the U.S. Army Medical Research and Development Command under Contract No W81XWH1920038.
The views expressed in this article are those of the author(s) and do not reflect the official policy or position of the U.S. Army Medical Department, Department of the Army, DOD, or the U.S. Government. Research was conducted in compliance with Animal Welfare Act, the implementing Animal Welfare regulations, and the principles of the Guide for the Care and Use of Laboratory Animals. The Institutional Animal Care and Use Committee approved all research conducted in this study. The facility where this research was conducted is fully accredited by the AAALAC.
AUTHOR DISCLOSURE AND GHOSTWRITING
Michael Broomhead and Elof Eriksson are employees of Applied Tissue Technologies that manufactures the XPansion kits. All the other authors declare no conflict of interest. No ghostwriters were used to write this article.
ABOUT THE AUTHORS
SUPPLEMENTARY MATERIAL
Supplementary Figure S1
Abbreviations and Acronyms
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
