Role of Ultrathin Skin Graft in Early Healing of Diabetic Foot Ulcers: A Randomized Controlled Trial in Comparison With Conventional Methods

With a rising prevalence worldwide, diabetes mellitus (DM) is a burden on the health care systems.¹ Diabetes mellitus is becoming more common; the number of people living with DM increased 1.5 fold from 463 million, and the prevelance is expected to reach 700 million in 2045.² After China, India has the second-highest population of individuals with diabetes.³ Diabetes is associated with multiple complications, including diabetic foot ulcers (DFUs).

According to the International Working Group on the Diabetic Foot (IWGDF), a DFU is a full-thickness wound passing through the dermis and located below the ankle in a patient with diabetes.⁴ It is estimated that 15% of patients with DM will experience a DFU in their lifetime⁵; if these patients are not treated in a timely manner, a DFU can progress to infection, osteomyelitis, or gangrene, which can lead to amputation. As many as 8 in 10 nontraumatic amputations can be attributed to DM, and 85% of such amputations result from DFUs.⁵ Mortality after unilateral limb amputation varies, with the risk between 13% and 40% at 1 year, between 35% and 65% at 3 years, and between 39% and 80% at 5 years.⁵ However, it has been estimated that the majority of these amputations can likely be prevented.⁶

Microangiopathy is one of the main pathologic conditions underlying the slow, insufficient, or nonhealing nature of a DFU.⁷ Other pathologic conditions include decreased or impaired growth factor production, collagen accumulation, and quantity of granulation tissue.⁸ Conventional treatment methods involve initial debridement and multiple dressings. Treatment could include repeat debridements and a long follow-up period until complete wound is achieved; long follow-up period is considered to exceed 3 months in this study. Many wounds take longer than expected to heal, resulting in an increased morbidity rate and health care costs. The estimated 5-year mortality is up to 30.65%, and the cost burden is equivalent to that of cancer.⁹

Standard therapy for managing DFUs involves glycemic control, antibiotic coverage, debridement of necrotic tissue, regular dressing changes, and offloading footwear. Several recent modalities have been introduced, such as various types of dressing, including negative pressure wound therapy (NPWT), offloading, growth factor application, and bioengineered skin constructs.¹⁰

Although NPWT shows promising results, it has disadvantages such as the need for multiple sittings, foam dressing fragmentation and retention, and difficulty maintaining effective suction. There may be a requirement for a definitive skin grafting procedure for coverage to encourage complete healing.^11,12 Hyperbaric oxygen therapy can enhance wound healing; however, it is not always readily available and may not be adequate for use in patients with large wound sizes involving more than 70% area of the foot, surrounding edema, and loss of tissue for local coverage.¹³ In addition, many complications have been associated with hyperbaric oxygen treatment. Topical growth factors can promote wound healing; however, it is costly and not readily available, and more conclusive studies are necessary to support the efficacy of such treatment.¹⁴ Split-thickness skin grafting (STSG) can be used to manage and close DFUs. Still, it can be used only when the wound bed is healthy, with sufficient granulation tissue, which in turn could demand multiple, regular saline-gauze dressings or NPWT applications.¹⁵ Often, even STSG may not be sufficient.

Conventional treatments for DFUs could be associated with more hospital visits and a longer healing time, which would add to the patient’s health care cost and potentially affect their overall quality of life. The cost involved, availability of the treatment, and frustration with the lack of healing can significantly affect patient adherence.¹⁶ Considering the socioeconomic circumstances of the population in India in particular, these factors may contribute to poor patient adherence to the treatment protocol, leading to further complications associated with DFUs.¹⁷

Galiano et al¹⁸ demonstrated the efficacy of topical vascular endothelial growth factor (VEGF) application on DFUs and observed accelerated wound healing through increased angiogenesis. Osborne et al¹⁹ demonstrated growth factor secretion ability in cell cultures of epidermal micrografts. They identified the presence of various growth factors such as VEGF and endothelial growth factor using immunohistochemistry. Based on these studies,^18,19 the authors of the current study proposed that epidermal micrografts might be used on DFUs, which could improve healing by initiating growth factor secretion at the wound bed. Such grafts might also act as a primary cover for wound closure.

In a prospective study, Hachach- Harametal²⁰ proposed a method of epidermal grafting using an automated tool (epidermal graft harvesting device; CELLUTOME Epidermal Harvesting System; 3M) in an outpatient setting, resulting in minimal or no pain and a scar-free donor site. A case series based in India demonstrated promising results of epidermal grafting as a potentially suitable option for managing chronic, complex wounds when only the epidermal layer is required.²¹ The epidermal skin graft has only been previously used on a wound in which the wound bed had sufficient granulation tissue formation (before multiple dressings or NPWT applications were used for wound bed preparation). To the authors' knowledge, epidermal skin grafts have not demonstrated the ability to secrete growth factors, which may result in better healing outcomes. With considering the above, the current authors explored using epidermal skin grafting (ultrathin skin grafting; UTSG) in the management of DFUs.

A newer treatment modality is needed for managing DFUs that addresses the problem of delayed healing by reducing follow-up time and treatment cost as well as being readily available. The purpose of the current study was to demonstrate the efficacy of UTSG in the early healing of DFUs in terms of cost-effectiveness, reduced total number of hospital visits, and a better final outcome of the wound (ie, limb salvage rate).

Trial design

This study was a single-center, randomized controlled trial with 2 parallel groups. Eligible patients were randomized to either the UTSG group or conventional dressing group using a computerized randomization method.

Research ethics approval

This trial received approval from the Institutional Ethical Committee and was registered at the Clinical Trials Registry-India. No.REF/2020/05/033393.

Study setting

Participants were recruited at the hospital on an outpatient basis.

Eligibility criteria

Patients referred by general surgeons for nonhealing DFUs were eligible for the study. Before enrollment, patients were screened for inclusion in the trial, and a patient information sheet was provided. This process included explaining the aims of skin grafting, methods of skin grafting and subsequent wound management, anticipated benefits, and potential hazards of the study. Patients were afforded sufficient time (≥ 24 hours) to consider whether they wished to participate. Patients were then invited to participate in the study, and informed consent was obtained. Treatment occurred within 7 days of patient enrollment. Patients were eligible for inclusion if they had an acute DFU of less than 3 months and could participate in the trial and adhere to the requirements, including weekly visits and the follow-up regimen. Patients were excluded if they had any of the following: chronic wounds of more than 3 months’ duration; Wagner grade III or IV DFUs; prior treatment with hyperbaric oxygen, NPWT, or growth factor application; and peripheral vascular disease (PVD) requiring vascular surgery intervention.

Clinical examination was routine, including obtaining a foot radiograph and a wound culture swab preoperatively. All wounds were assessed clinically, and wound size and quantity of granulation tissue were calculated using Mobile Wound Analyzer software (HealthPath) based on photographic images taken by a single camera for all cases. The Pressure Ulcer Scale for Healing (PUSH) tool was used to assess wound severity.

Routine lower limb palpation was carried out to look for dorsalis pedis artery, anterior tibial artery, and posterior tibial artery pulsations; cases with feeble or absent pulsation were advised to receive arterial Doppler to look for PVD. Patients with absent or monophasic Doppler flow requiring urgent vascular surgical intervention were excluded from the study. Patients with adequate flow and for whom a vascular surgeon advised conservative treatment were included in the study. Patients in the test group underwent debridement and UTSG in a single session.

Ultrathin grafting

Prior to grafting, the wound was debrided, necrotic patch and slough were excised, the wound edge was freshened, and thorough saline irrigation was performed. In addition to the wound being completely debrided, thorough irrigation was performed after complete hemostasis had been achieved.

As is typical, the graft was taken from the opposite thigh. The donor site was cleaned, prepped, and draped. Saline was injected into the subepidermal plane using a 10-mL syringe to separate the superficial epidermis and deeper dermis. The amount of saline injected depends the graft size required. Emollient liquid paraffin was applied on the top to facilitate easy gliding of the blade. The Watson modification of the Humby knife with the lowest blade setting was used for graft harvesting (Figure 5). To confirm the UTSG, the ultrathin graft was macroscopically identified by its transparent nature and punctuated minor bleed at the donor site (Figure 6). The graft was applied over the wound and fixed with multiple staples to ensure contact between the graft and the wound bed. The wound was dressed with petrolatum gauze, followed by applying a nonadherent soft pad with a soft cloth backing. An 8-layer plaster of Paris slab was used to immobilize the ankle joint. The donor site was covered with petrolatum gauze, and a tight compression dressing was applied. The wound dressing was first opened on postoperative day 5 and subsequently as required (usually every third day). Staples were removed after 10 to 14 days.

Wound bed biopsy

Incisional tissue biopsy at the center of the wound bed was performed before grafting and 3 weeks after grafting. Biopsy prior to grafting was done under spinal anesthesia, and a repeat biopsy was performed at week 3 after grafting following administration of adequate local anesthesia (2% lidocaine). The specimens were placed in a sterile vial containing 4% formaldehyde and transferred to the laboratory for histopathologic examination.

Conventional dressing

The control group underwent initial debridement and multiple regular saline-gauze dressings. Local debridement of slough on an outpatient basis was done at each visit if required. Patients were observed for progression to abscess or osteomyelitis, and the presence of either resulted in immediate debridement or amputation.

Study outcome

The primary endpoints were complete healing of the wound after grafting or completion of the study period (12 weeks). Secondary endpoints included wound-related adverse events (eg, progression of infection) leading to interventions such as amputation during the study period.

Wound healing was assessed clinically, with wound measurement recorded using the wound analysis software at each visit (Figure 7). High-quality, accurate, standardized images for digital measurement of the wound surface area were obtained and stored in the digital photograph diary.

Study outcomes measured were time to wound healing, total number of hospital visits and cost of treatment, final outcome of the wound, and the incidence of adverse events occurring within the duration of the study. Furthermore, the authors determined the wound-healing mechanism of UTSG by analyzing histopathologic reports of the wound bed tissue biopsy to assess improved granulation tissue and vascularity of the wound bed (Figure 8).

Participant timeline

The study was opened to recruitment in January 2017 and closed in September 2019. Each patient was to receive weekly follow-up visits for 12 weeks or until wound healing occurred. The final review was to occur at the end of the third month following treatment initiation (ie, at the last hospital visit).

Sample size

The present pilot study was given a significance level of .05 for 80% power, yielding a sample size of 26 patients per group. The goal was to recruit a total of 52 patients into the study.

Randomization, allocation concealment, and blinding

Once consent was obtained, patients were randomly assigned to either the test group or the control group. A random allocation sequence was computer generated using SPSS Statistics for Windows, version 22.0 (IBM Corporation). The allocation sequence was sealed in identical opaque envelopes and given to the enrolling researcher upon receipt of patient consent.

Blinding

The surgical team, clinical staff, and patients were not blinded to the intervention status. Independent blinded analysis of the photographic diary was carried out by 2 plastic surgeons (other than the operating surgeon); they were blinded to the intervention and study groups. Outcome parameters were recorded as per their inference.

Data collection

All data collected were recorded on paper forms and in a digital folder. The surgical team and trial personnel collected data. A research fellow ensured the accuracy of the data collection by performing sample assessments at regular intervals. Any adverse events were recorded and reported to the primary investigators as well as the institutional ethics committee. Wounds were assessed and recorded in a wound assessment form at each visit. The wound analysis software applied to photographic images was used to measure the wound surface area digitally. The number and cost of outpatient visits were recorded, and the type and cost of the dressings used were documented.

Calculation of the total cost of treatment included the cost of hospitalization as well as the procedure charge and dressing charges at each subsequent hospital visit.

Statistical analysis

Descriptive and inferential statistical analysis was performed in the current study. Results of continuous measurements are presented as the mean, and the results of categorical measurements are presented as n (%). A P value less than .05 was considered statistically significant. The following assumptions about data were made: (1) dependent variables should be normally distributed, (2) samples drawn from the population should be random, and (3) cases of the samples should be independent.

The Student t test (2-tailed, independent) was used to calculate the significance of study parameters on a continuous scale between 2 groups (intergroup analysis) on metric parameters. Either the χ² test or the Fisher exact test was used to calculate the significance of study parameters on a categorical scale between 2 or more groups; nonparametric setting was used for qualitative data analysis. The Fisher exact test is used when cell samples are very small.

Statistical analysis was performed using SPSS Windows, version 18.0 (IBM Corporation) and R version 3.2.2 (Bell Laboratories). Word and Excel (Microsoft) were used to generate graphs and tables.

Patients who received the study treatment were evaluated for analysis. If the clinical course could not be fully evaluated, the date of the last visit was considered to be the final data point for analysis. Baseline characteristics of the 2 groups were recorded. Mean time to wound healing was determined based on the number of weeks until complete reepithelialization.

A total of 56 patients were enrolled in the study. Of these, 3 were lost to follow-up before 12 weeks, and 1 patient received collagen dressing at the local clinic. Thus, 4 patients (2 from each group) were ineligible for inclusion. Data from 52 patients (26 in each group) were analyzed. Age distribution among both groups was statistically matched with the mean age in the test groups (56.92 and 57.79 years; P =.44, Student t test) (Table).

Sex distribution was matched, with 11 males and 15 females in the test group and 14 males and 12 females in the study group (P =.405 [moderately significant] per χ² test).

Among 52 patients, 4 had associated PVD and were undergoing conservative management as per the advice of the vascular surgeon. Two of these patients were in the test group, and 2 were in the control group.

Wound severity was measured using the PUSH tool. Before intervention, the mean wound score was 14.42 in the test group and 14.27 in the control group, which was statistically comparable. Before intervention, the mean wound size was 54.85 cm² ± 47 in the test group and 39.38 cm² ± 17.58 in the control group (P =.016). At week 6, the mean wound size had decreased to 25.08 cm² ± 39.29 in the test group and 28.69 cm² ± 17.56 in the control group (P =.617). At week 12, the mean wound size had further reduced to 9.08 cm² ± 27.62 in the test group and 27.68 cm² ± 16.88 in the control group (P =.001) (graphical representation is shown in Figure 1).

At week 6, there was a 54.50% reduction in mean wound size observed in the test group, and a 27.14% reduction in mean wound size noted in the control group. At week 12, the reduction in mean wound size was 83.44% in the test group and 29.68% in the control group compared with the preintervention wound size (Figure 2, Figure 3).

Complete wound healing was achieved in 22 of 26 patients (84.61%) in the test group and 14 of 26 patients (53.84%) in the control group during the 12-week study period. The mean time to heal was 6.22 weeks in the test group.

The mean number of hospital visits was 9.2 in the test group and 11.07 in the control group. Four patients in the test group required more than 12 visits (15.38%), and 13 patients in the control group required more than 12 visits (46.15%).

The total cost of the intervention was less than 15  000 INR ($198 US) for 11 patients in the test group and 9 in the control group, 15 000 INR ($198 US) to 30 000 INR ($397 US) for 9 patients in the test group and 5 in the control group, and more than 30 000 INR ($397 US) for 6 patients in the test group and 12 in the control group (P =.001 [Fisher exact test]).

The final outcome of complete wound healing was achieved in 22 wounds in the test group and 14 wounds in the control group (84.61% and 53.84%, respectively) (Figure 4). Among the remaining wounds, 1 resulted in amputation in the test group, whereas 3 in the control group underwent amputation. During the 12-week study period, 3 wounds in the test group did not heal, and 9 wounds in the control group did not heal (P =.009 [Fisher exact test]).

Complete graft loss was noted in 2 patients, which was expected owing to association with PVD. However, in these cases, the present authors observed clinical signs of wound healing (ie, increased granulation tissue and hyperemia), which allowed repeat grafting with STSG. Later wounds healed completely without any complications. From the results of the current study, the ultrathin graft did not persist in the wounds but seemed to have a long-term effect on tissue regeneration. It appears the ultrathin graft is essential for initiating tissue repair but is dispensable after the patient’s cells are activated.

No significant complications were noted at the graft harvest site wound. Two patients had delayed spontaneous healing.

Histopathologic examination of a wound bed tissue biopsy taken before intervention and 3 weeks after UTSG showed a comparative increase in granulation tissue and tiny blood vessels in the test group (Figure 9, Figure 10, Figure 11, Figure 12, Figure 13, Figure 14, Figure 15, Figure 16, Figure 17).

Wound care for those with diabetes is a significant financial and resource burden on the health care system, which signifies a need to further optimize current wound coverage strategies.^22,23 Epidermal grafting for wound healing is not a new concept. Several case reports have indicated good wound healing outcomes; however, it is unknown whether the healing rate is comparable to that of STSG, a mainstay of treatment for wounds that cannot be closed primarily.^24-26 The usefulness of a novel epidermal harvesting system was first demonstrated in Port-au-Prince, Haiti, where insufficient resources and lack of clinical training limited wound care options.²⁷

The use of epidermal graft in managing vitiligo and chronic wounds has been widely reported, but its use is limited because of the lack of reproducible and efficient harvesting techniques.^28,29 The present study demonstrated the feasibility of using the Watson dermatome for harvesting UTSG to achieve definitive wound coverage of DFUs.

Similarly, automated epidermal graft also proved useful when STSG was contraindicated because of concern for poor wound healing at the donor site, such as in the case of pyoderma gangrenosum.²⁶ Molecular studies of the epidermal graft using an automated epidermal harvesting tool showed that epidermal micrografts formed at the dermal-epidermal junction secrete various growth factors essential to wound healing, including platelet-derived growth factor, VEGF, and granulocyte colony-stimulating factor,^19,30 thereby encouraging the wound bed to regenerate and initiate keratinocyte migration from the edges of the wound. The migrating keratinocytes also deposit a variety of extracellular matrix components, such as laminin, fibronectin, and type IV collagen.³¹

These findings are consistent with the observation of the authors of the current study that when the epidermal graft was transplanted, wound healing occurred simultaneously within the wound bed and margins. These islands of reepithelialization eventually merged to form a confluent structure, which suggests a unique molecular mechanism of epidermal grafts and warrants further investigation.

Wound healing and healing time were key outcomes measured, and the results demonstrate that 84.61% of the wounds in the test group fully healed. More than 50% of wound healing was achieved within 6 weeks, and complete wound closure was achieved within a mean of 6.2 weeks. Of the wounds that were not healed in the study population, the majority (75%) were in the control group, which may imply that the epidermal grafts stimulate the healing process in hard-to-heal wounds.

Ultrathin skin graft donor site complications can be avoided, such as infection, pain, and hypertrophic scarring.³² None of these donor site-related complications occurred in the current study.

As with all new technologies, the costs of intervention and follow-up time must be assessed; these factors may also be important determinants of patient adherence. In the present study, the mean frequency of hospital visits was 9 among 84.6% of the study population in the test group, whereas 50% of the patients in the control group required more than 12 visits. This would mean patients in the control group are associated with a higher number of hospital visits, thus adding further health care costs to the group that did not receive the UTSG. The probable reason for it may be that the extra added cost of UTSG is compensated by multiple dressings, more hospital visits for a longer period, and more cost to the patients in the control group.

In the current study, the effect of the growth factors on the wound bed and the edges of the wound was further confirmed by histopathological examination. The incisional biopsy taken from the center of the wound was used to identify increased granulation tissue and tiny blood vessels, which are signs of wound healing probably due to a defect of growth factor.

Also, in patients with DFUs with mild PVD being managed conservatively, the current study demonstrated the efficacy of UTSG by improving wound healing in terms of improved granulation tissue and hyperemia of the wound bed. The primary aim was to demonstrate an improved wound bed, which was easily accomplished, after which the wound was covered with STSG, and complete wound closure was achieved.

Ultrathin skin grafting serves an important role in improving the healing process of the wound bed and acting as a primary cover for wound closure in most cases. Some cases may require additional STSG for definitive wound closure, which shows promising results due to enhanced healing process by prior UTSG. In the authors' opinion, most wounds healed with UTSG, but in some cases, an additional STSG may be necessary. However, the use of UTSG prior to definitive STSG would help to increase granulation tissue and help with STSG uptake.

Ultrathin skin grafting has advantages over STSG; UTSG allows contact between the growth factor-secreting basal epidermal layer and the wound bed and improves the healing process, which is not exposed as it is in STSG. Ultrathin skin grafting was associated with fewer harvesting site complications than other various modalities used in the treatment of DFUs, which was a consistent finding in the current study.

This study has several limitations, including small sample size and a short study period. Although UTSG secretes growth factor, this investigation did not measure the actual values via immunohistochemistry; however, it should be considered in future research regarding UTSG. Additionally, it lacks matching of confounding factors that might alter the healing process, such as nutrition, smoking, alcohol consumption history, personal hygiene, physical activity, neuropathy, and PVD.

Ultrathin skin grafting offers early healing of DFUs owing to its growth factor–secreting ability and acts as a primary cover for wound closure. It also requires fewer hospital visits and has a lower total treatment cost compared with conventional dressing treatment. Ultimately, UTSG improves the final outcome of the wound. It is a more straightforward technique than other recent treatment modalities, has a short learning curve, and requires no special equipment other than standard instruments. Cost-effectiveness and availability promote patient adherence to treatment, which makes up for poor follow-up with other treatments.

Ultrathin skin grafting offers a new modality in managing DFUs, which is more feasible than conventional frequent dressing changes. Additional studies are required to prove the present authors’ hypothesis.

Authors: Rahul Shetty, MBBS, MS, MCh, DNB¹; BS Giridhar, MBBS, MS²; and Ankush Potphode, MBBS, DNB³

Affiliations: ¹Department of Plastic Surgery, St. Martha’s Hospital, Bangalore, India; ²Department of General Surgery, Bangalore, India; ³General Surgery, St. Martha’s Hospital, Bangalore, India

Correspondence: Rahul Shetty, Department of Plastic Surgery, St. Martha’s Hospital, Banglore, Karnataka, India 560003; rahulplastic@hotmail.com

Disclosure: The authors declare that they have no conflicts of interest.

Recommended Citation: Shetty R, BS Giridhar, Potphode A. Role of ultrathin skin graft in early healing of diabetic foot ulcers: a randomized controlled trial in comparison with conventional methods. Wounds. 2022;34(2):57-67. doi:10.25270/wnds/010522.01