Human Skin Allograft for Patients With Diabetic Foot Ulcers, Venous Leg Ulcers, or Surgical/Traumatic Wounds: A Retrospective, Descriptive Study

Chronic wounds such as diabetic foot ulcers (DFU) and venous leg ulcers (VLU) may take a long time to heal and increase the risk of complications. Previous studies have suggested human skin allograft may facilitate healing of these chronic wounds. A retrospective, descriptive study was conducted among outpatients with nonhealing DFU, VLU, surgical, or traumatic wounds managed with a meshed, partial-thickness, cryopreserved human skin allograft. Charts of all patients who received an allograft from 2011 to 2013 were abstracted if the wound was >1 cm2, had a duration >30 days, was adequately debrided, and was free of infection before the first allograft application. Primary outcome was percentage of wounds healed (ie, 100% epithelialized) at 12 and 20 weeks. Secondary outcome was the number of recorded adverse events. Wound measurements (area — calculated as width x length in cm2), wound type and duration, number of allograft applications, number of adverse events, and race, smoking status, and body mass index were abstracted. Of the 49 patients (average age 64.3 [SD 15.0]; 64% male) who met the inclusion criteria, 13 did not have medical follow-up through the primary outcomes at 12 and 20 weeks, leaving 36 patients (average age 65.1 [SD 15.4]; 67% male) available for analysis. The most common diagnoses were VLU (18 patients, 50%) and traumatic wounds (9 patients, 25%). Average wound size was 19.4 cm2 (SD 29.3, range 1.2–156, median 9.5), and average wound duration at initial treatment was 17.2 (SD 17.0, range 4–72, excluding outlier) weeks. Seventeen (17) wounds (47%) healed by 12 weeks, and 21 (58%) were healed by week 20 with an average of 3.3 (SD 2.0) allograft applications. No serious adverse events occurred. The results of this study are encouraging and add to the currently available literature on the use of allograft skin for chronic wounds, but the study design and sample size limit the ability to interpret the observations. Prospective, controlled clinical studies are needed to compare the efficacy, effectiveness, and cost-effectiveness of human skin allograft to standard care and to other advanced care modalities.

A review of the literature suggests an estimated 1% to 2% of the population will be affected by leg or foot ulceration, which often is associated with diabetes mellitus or venous disease.¹ Such ulcers can become chronic wounds when they do not heal in an orderly and timely manner.² In addition to reducing quality of life, chronic wounds result in significant health care expenditures.^3-7 The incidence of chronic wounds in the population is likely to increase with the increasing prevalence of obesity and diabetes.^8-10

Using the 2005–2008 National Health and Nutrition Examination Survey, the 2007–2009 National Health Interview Survey, 2009 Indian Health service data, and 2010 US resident population estimates, the United States Centers for Disease Control and Prevention¹¹ estimates 6.0% of the US adult population (more than 18 million people) have been diagnosed with diabetes and an additional 7 million live with undiagnosed diabetes. Diabetic neuropathy and impaired blood circulation, common complications of diabetes, contribute to foot deformities and ulcers.¹² More than 12% of adults with diabetes have a history of foot ulcers, which is associated with an increased risk for lower extremity amputation (LEA).¹³ In 2006–2008, the yearly prevalence of DFUs among the Medicare population with diabetes was estimated to be 8.0% and the yearly prevalence of LEA in the same population was estimated to be 1.8%.¹⁴

DFUs add to the already high cost of diabetes care.¹⁵ The results of a prevalence-based study using the 2012 US population showed the incremental cost of treating DFUs is $11,700 to $16,883 per patient (2012 US$), between $9 to $13 billion annually in the US.⁷

Venous leg ulcers (VLUs) may result from venous incompetence and venous hypertension.¹⁶ Primary risk factors for VLUs are increasing age, obesity, previous leg injuries, deep venous thrombosis, and phlebitis.¹⁷ In 2011–2012, the prevalence of adult obesity in the US reached 35%.¹⁸ Results of a retrospective, cohort study¹⁹ using data from 1988–1996 suggest the estimated annual prevalence rate of VLUs among persons age 65 is 1.69%, with an annual incidence rate of 1.13 to 1.20 per 100 persons. A comprehensive review²⁰ reported prevalence estimates of active or healed venous ulcers in the general population range from <1% to 5%, depending on sampling methods and the population studied. Health care costs generated in the treatment of VLUs are high. Although newer estimates are wanting, a 1997 retrospective cohort study⁶ estimated VLU treatment costs at $9,600 per patient in the US; after 2013 Medical-CPI adjustments,²¹ that amount is $17,000 per patient.

Standard wound care for DFUs typically includes debridement, management of moisture balance, and offloading to relieve pressure. If needed, advanced treatments such as therapeutic growth factors, bioengineered tissues, hyperbaric oxygen therapy, or negative pressure wound therapy (NPWT) can be employed.²² VLUs are typically treated with compression therapy and moisture-retentive dressings as well as the advanced treatments noted for use with DFUs when wound healing does not progress.^16,17

Effective treatments for chronic wounds are needed to ease the high patient and health care cost burdens. A growing body of evidence from retrospective studies, randomized prospective studies, and case studies suggests the application of human skin allograft in the form of cryopreserved cadaveric skin may improve healing of chronic wounds compared to standard wound care or bioengineered skin substitutes.^23-27

AlloSkin™ (AlloSource®, Centennial, CO), hereafter referred to as AHSA, is a meshed, partial-thickness, cryopreserved human skin allograft processed from donated cadaveric tissue. The donated tissue undergoes extensive serological and microbiological testing for specific micro-organisms and fungal contaminants and is evaluated to ensure graft acceptability. AHSA has been found through proteomic analysis²⁸ to contain extracellular matrix proteins, glycosamnoglycans, cytokines, collagens, and other natural compounds. Collagen is known to aid in wound healing by guiding fibroblasts that migrate along the connective tissue matrix; the collagen fibers attract fibrogenic cells that are essential to wound healing.

AHSA is regulated by the US Food and Drug Administration (FDA) as a Human Cell, Tissue, and Cellular Tissue Based Product (HCT/P) under Code of Federal Regulations Title 21 Parts 1270 and 1271²⁹ for use in the care of adequately debrided wounds that are free of infection, including burn wounds and nonburn acute and chronic wounds. After thawing in sterile isotonic solution for a minimum of 1–5 minutes, AHSA is temporarily secured on the wound bed with adhesive strips, surgical glue, or tacking with sutures. According to the manufacturer,³⁰ the allograft tissue generally sloughs in 7–14 days as granulation tissue develops. Additional applications are used if satisfactory granulation tissue formation has not been achieved with the initial application.

Human skin allograft is standard therapy for patients with extensive burns.^23,31 It functions as a protective barrier and provides collagen and structure to the wound bed, factors that may facilitate healing of other types of wounds. A recent retrospective review²⁷ described the use of allograft as a biologic dressing for chronic ulcers, DFUs, necrotizing fasciitis, and acute traumatic wounds; histological analysis showed evidence of tissue repair in the migration of epithelia from the patient’s skin to the surface of the allograft and the presence of granulation tissue at the base of the allograft.

The purpose of this retrospective descriptive study was to evaluate wound care outcomes and the safety of using AHSA in the care of patients with chronic DFU, VLU, surgical, or traumatic wounds. 

Design and setting. This retrospective observational chart review was conducted among patients treated with AHSA from July 2011 to February 2013 at the wound clinics of the Inova Health System, a not-for-profit health care system based in Northern Virginia. These facilities include Inova Mount Vernon Hospital, Inova Fair Oaks Hospital, and Inova Wound Healing Center – Annandale. AHSA is used regularly at the participating Inova sites to treat a variety of patients. The clinical study protocol, Retrospective Clinical Study of AlloSkin Use in Subjects with Diabetic Foot Ulcers (DFU), Venous Leg Ulcers (VLU), or Surgical/Traumatic Wounds, allowed for review of up to 220 patient records to allow for exclusion of patients with adjunctive advanced treatment in conjunction with AHSA. Institutional Review Board (IRB) approval was granted at the Inova Health System. The study was performed in accordance with all applicable privacy laws and human research protections. All data were considered confidential and patients were de-identified, receiving a study identification number that was used during data review and analysis.

Study population. Patients with a DFU, VLU, surgical, or traumatic wound treated with AHSA were eligible for inclusion if the initial wound size was >1 cm² and the wound duration was >30 days before to the first application of AHSA. Patients were excluded if gross infection at the wound site was noted in the medical chart or additional treatment was received after the initial AHSA treatment. Additional treatments that led to patient exclusion included hyperbaric oxygen treatment, venous ablation procedure, lower extremity bypass graft, angioplasties or stents, and additional growth factor applications, as well as the use of biological products containing cells or other tissue-derived products, matrix components, or enzymatic debridement agents. The adjunctive use of NPWT, compression therapy, or use of offloading did not exclude patients from the study.

Variables and outcomes. Medical charts were reviewed through a clinical research organization designated by the study sponsor. The charts were abstracted by a clinical research associate (CRA) trained and experienced in clinical trials. The data were recorded in a digital spreadsheet following IRB-approved protocol and applicable Good Clinical Practice guidelines. Data collected from the medical charts included wound length and width measurements, type, location, and duration; and age, gender, race, body mass index (BMI), and smoking status. Additional data collected included number of allograft applications, number of adverse events (AE), and demographic information.

Patient medical records were reviewed for a maximum of 20 weeks, ending earlier if the wound closed, an infection required surgical incision and drainage, or the chart was incomplete. Data collection and abstraction frequency were not coordinated with patient visit frequency. Data were abstracted biweekly from the time of the first AHSA application up to 12 weeks and for week 16 and 20 thereafter. Primary wound care outcomes were the percentage of patients healed at 12 and 20 weeks after the first AHSA application. Healing was defined as a notation of 100% epithelialization. Secondary study outcomes were the number of recorded AE related to AHSA or the target ulcer that occurred between the initial application of AHSA and up to 20 weeks thereafter. AE were recorded as nonserious AE related to AHSA, serious AE related to AHSA, nonserious AE related to the target ulcer, or serious AE related to the target ulcer. The CRA determined the type of AE using information in the chart.

Data analysis. Descriptive statistics were prepared using SAS version 9.4 (SAS Institute, Inc., Cary, NC). Wound size (area) was calculated as wound length multiplied by width, each measured in cm. Mean, median, standard deviation, and range were analyzed for continuous variables. Frequency and percentage were reported for categorical variables.  

Of the 49 patients who met the initial inclusion criteria, 36 (24 men, 12 women, average age 65 [range 22–94, median 67] years, each with 1 wound) had sufficient follow-up duration data for analysis. The 13 patients unavailable for analysis included 10 patients lacking medical follow-up data for the primary outcome at 12 and 20 weeks and 3 that died unrelated to the study. All patients received outpatient care. Average BMI was 30.6 (range 16.8–67.6, median 28.0). Of the 36 wounds, 18 (50%) were VLUs, 9 (25%) were traumatic wounds, 7 (19%) were DFUs, and 2 (6%) were surgical wounds. Average wound duration before initial AHSA treatment was 70.2 [range 4–1,924, median 11] weeks. After excluding the extreme value, the average duration was 17.2 [range 4–72, median 11] weeks; 11 (30%) had wounds with duration of 20 weeks or more before the first AHSA application. Average wound size was 19.4 cm² (range 1.2–156, median 9.5), also influenced by an extremely large wound measuring 156 cm². The average after excluding the extreme value was 15.5 cm² (range 1.2–71.4, median 7.2). Fifty percent (50%) of wounds were >10 cm² at baseline, and 30% were 20 cm² or more (see Table 1).

Of the 36 wounds, 17 (47%) were healed by 12 weeks and a total of 21 (58%) healed by 20 weeks. The 20-week healing data by wound type included 5 out of 9 (56%) traumatic wounds, 11 out of 18 (61%) VLUs, and 5 out of 7 (71%) DFUs. Neither of the 2 surgical wounds healed by 20 weeks. The rate in the number of patients healed was highest through week 10, continuing at a slower rate through week 20 (see Figure 1). For wounds that healed by 20 weeks, average time to closure was 9.2 weeks (range 4–20 weeks, median 8 weeks) (see Table 2).

Patients received an average of 3.3 AHSA applications (range 1–8, median 3), including the initial application. Patients that healed by 20 weeks received 2.7 AHSA applications (range 1–6, median 2) compared to 4.0 applications (range 1–8, median 3) for patients that did not heal. The majority of patients received between 1 and 3 applications of AHSA (see Figure 2). Among patients that healed by 20 weeks, 16 out of 21 received 3 or fewer applications.

No serious or nonserious AE related to AHSA use occurred. The 3 deaths were unrelated to the target ulcer or the use of AHSA. Two nonserious AE were related to the target ulcer, both indicated in the medical record as pre-study, patient-reported chronic pain.

In this retrospective review of chronic wounds treated with AHSA, 58% of wounds healed by 20 weeks. Before AHSA treatment, the wounds had an average duration of 70 weeks and an average size (baseline area) of 19.4 cm² (17 weeks, 15.5 cm² excluding outliers).

Human skin allografts have a number of properties that promote healing. Allografts have been found in histological studies²³ to increase vascularity, decrease bacterial contamination, and stimulate epithelialization. A comprehensive review²³ presented evidence that human skin allografts stimulate wound granulation tissue act as a barrier to bacterial invasion;  trigger immune response to reduce bioburden; decrease water, electrolyte, and protein loss; prevent desiccation of substructures; stimulate neovascularization; and diminish pain.

Retrospective²⁵ and randomized studies²⁶ have reported healing rates of 67% to 75% at 20 weeks after the first application of a human skin allografts. In the largest of those studies²⁵ (N = 188), a retrospective study of patients with VLUs (N = 134) and DFUs (N = 54), average baseline wound size was 6.2 (±11.8) cm² in the DFU group and 11.8 (±22.5) cm² in the VLU group. Wound duration averaged 18.7 (±28.8) weeks before the initial allograft application. An average of 3.23 (±2.77) allograft applications was applied by the 20-week point. The second study,²⁶ a randomized trial, included a much smaller number of patients with a DFU (N = 12) that were randomly assigned to allograft therapy; in this study, average wound size at baseline was 1.82 cm² (wound duration before allograft application was not reported), with an average of 1.38 (±0.29) allograft applications. The healing rate from the current retrospective study was somewhat lower, which might be explained by baseline wound size. Larger wounds require a longer time to heal, and the odds of healing by 12 weeks decreases as the wound size increases.²⁵ Average baseline wound size in the current study was much larger than the other reported studies (ie, 19.4 cm² compared to 1.82 cm² and 10.2 cm², respectively,^25,26 even after removing the extreme value [15.5 cm²]). This is the first known study of allograft use on wounds of this size.

An alternative advanced therapy in the wound care armamentarium is living cell-based skin substitutes, which have been evaluated for the treatment of DFUs and VLUs. In randomized controlled trials,^32-35 12-week healing rates for the products ranged from 30% to 56%, with average wound sizes of 2.3 cm² to 6.6 cm²; only 1 study³⁴ included wounds >20 cm². The healing rate from the current study — 47% at 12 weeks — compares favorably to the healing rates of living cell-based skin substitutes, especially considering 30% of the wounds in the current study were 20 cm² or larger.

The cost of advanced wound therapies is a concern; however, therapies may be cost-effective if they improve healing rates, shorten healing times, and avoid serious and expensive complications, such as amputation.^6,36 In a 2003 cost-effectiveness study, Redekop et al³⁶ compared the 12-month costs of advanced wound care (living cell-based skin substitute, which includes growth factors and cytokines also present in AHSA, that has been shown in a randomized controlled trial³³ to improve healing of DFUs) to standard care as previously described for DFUs. The authors concluded the higher cost of advanced care was offset by avoidance of serious AE and resections or amputations.³³ Their cost-effectiveness analysis compared therapies, evaluating the additional expenditure required for each unit of improvement (ie, the product cost component of advanced therapies is relevant for determining cost-effectiveness).

A broad range of product prices has been observed, with AHSA falling at the lower end of this spectrum.^37-40 The cost per application of skin substitute ranged from $1,500 (in 2008)³⁹ to $1,664 (in 2011).³⁸ The products were reimbursed at $1,371 by the Centers for Medicare and Medicaid Services in 2013.⁴⁰ In contrast, AHSA cost per application is about one third of those therapies.³⁷ Lower product price and a comparable healing rate suggest a cost-effectiveness analysis of AHSA would yield favorable results.

To date, only 1 small randomized trial²⁶ has been conducted to compare 2 advanced therapies: patients with DFUs (N = 28) of wound size <2 cm² were provided either a bioengineered skin substitute (N = 16) or allograft (N=12) to evaluate wound closure. This study found 47% wound closure at 20 weeks for the bioengineered skin substitute compared to 67% for allograft. The results of the current study suggest prospective, randomized clinical studies to compare chronic wound outcomes of AHSA managed wounds to standard care and other advanced wound treatment modalities are warranted.

This study is limited by its retrospective design, small sample size, and lack of a comparison group. To evaluate the impact of AHSA, patients were excluded if they received additional treatment after the initial AHSA application. The exclusion was necessary to measure the impact of AHSA without biasing the results from use of adjuvant treatments, but it significantly reduced the pool of patients available for the study because adjunctive therapy is commonly utilized at this institution. Ten (10) patients met the inclusion criteria but were not included in the analysis because they did not continue care through the primary endpoints of 12 and 20 weeks, which may have biased the results. The inclusion criteria allowed for patients with DFUs, VLUs, surgical, or traumatic wounds that met wound duration and size criteria. However, that resulted in wounds of various types, which also limits the conclusions that can be drawn. In addition, the retrospective study design precluded control of additional variables that may have affected outcomes before or during AHSA care, such as offloading, compression, debridement, glucose control, or concomitant use of NPWT.

A retrospective study that evaluated the percentage of patients with chronic DFU, VLU, surgical, or traumatic wounds healed at 12 and 20 weeks when treated with AHSA found 21 of the 36 wounds (58%) were healed after 20 weeks. On average, 3 AHSA applications were required. This study adds to the growing body of literature on the use of allograft skin in the wound healing armamentarium by including patients with large (mean 19.4 cm²; 15.5 cm² when outlier is excluded) chronic wounds. Prospective, controlled clinical studies are needed to compare the efficacy and effectiveness of AHSA to standard care and to other advanced care modalities.

The authors thank the Inova Wound Healing Center, Inova Research Center, and especially Tilani Lowman, PA-S and Joan Wirsing, CCRN CHRN CWS for their assistance in this research study.

This study was funded in part by AlloSource®, Centennial, CO, with whom Dr. Anderson has a consulting arrangement, performing work for hire.

Dr. Desman is a Medical Director, Inova Wound Healing Centers, Alexandria, VA. Mr. Bartow is a retired Clinical Manager, Inova Wound Healing Centers, Alexandria, VA. Dr. Anderson is a Principal Health Economist, Technomics Research LLC, Bloomington, MN. Please address correspondence to: Louise H. Anderson, PhD, Technomics Research LLC, 8505 Penn Circle, Bloomington, MN 55431; email: landerson@technomicsresearch.com.