Use of Amniotic Tissue–Derived Allografts Post-Mohs Micrographic Surgery: A Preliminary Study Assessing Wound Closure Rate

After skin cancers are removed by Mohs micrographic surgery, wounds in locations such as pretibial skin, lower extremities, scalp, conchae, and digits are typically left open and may require several weeks to close, potentially necessitate multiple follow-up visits, and can carry a significant risk of infection. This study investigated the effectiveness of amniotic tissue–derived allograft (ATDA) to improve outcomes in post-Mohs wound care in areas where wound closure is difficult, focusing on time to wound closure and assessment of cosmetic outcomes.

Amniotic tissue–derived allograft has been shown to improve wound healing in a variety of clinical settings. Numerous studies have documented the effectiveness of ATDA in improving wound healing in diabetic foot wounds refractory to standard treatment modalities.^1-15 Although previous studies have demonstrated that weekly applications of membrane over the wound area have resulted in increased rates of wound healing,^1,2,11,15-17 the properties of amniotic fluid have been less well studied.

Preparations of ATDA have been shown to contain cytokines, growth factors, protease inhibitors, angiogenin, and extracellular matrix (ECM) components, which are extracted from the fluid and released from the membrane to the wound, promoting cell proliferation and migration in vitro.^18-21 In addition, as a scaffold rich in ECM, amniotic membrane has been shown to effectively bind a variety of cell types,^20,22-24 is thought to mobilize endogenous growth factors and cytokines,^20,25,26 and appears to increase host cell infiltration.^26-28 Thus, the suspected mode of action of wound healing is to release growth factors and cytokines,^18-21 attracting stem cells and promoting both angiogenesis and stem cell differentiation.^20,25,26,28

A preliminary case series suggests that the use of dehydrated amnion/chorion membrane allograft has shown efficacy in full-thickness defects after Mohs micrographic surgery¹⁶; however, to the authors’ knowledge, the efficacy of this allograft in secondary intention wound healing has not been systematically investigated. If beneficial effects can be demonstrated in the post-Mohs setting, improved patient outcomes may include decreased time to wound closure and decreased postoperative complications. Benefits to the health care system may include a decreased need for skin grafts, decreased number of postoperative visits, and an overall reduction in health care costs.

Institutional Review Board approval was obtained, and all patients enrolled voluntarily provided consent after being fully informed of the risks, potential benefits, and study protocol. Children and adults with incapacity were excluded from this study. Persons with diabetes and those with immunosuppression were also excluded to control for reduced healing times. 

Patients who met the study criteria were asked to volunteer to participate in this study and, if interested, participate in the informed consent protocol. No distinction was made between patients with different skin cancer diagnoses, and volunteer patients were enrolled identically independent of initial diagnoses. Patients assigned to the liquid treatment group were given one injection of PalinGen Flow (Amnio Technology) at multiple sites around the wound perimeter, with the volume of liquid scaled to wound size of 1 mm/cm². In the membrane (dry) treatment group, a size-adjusted overlay of PalinGen Membrane (Amnio Technology) was used; the membrane was reapplied at each follow-up visit. Standard dressing changes, which were applied to all patients, included sterile Vaseline Xeroform petrolatum (DeRoyal) followed by Tegaderm gauze (3M), sterile gauze, and Hypafix tape (Smith+Nephew). At follow-up visits, silver nitrate was applied to wounds in which the physician deemed hypergranulation tissue to be present. Patients were instructed to keep wounds covered and dry. They returned to the clinic for frequent bandage changes if the bandage was damp or needed replacing. 

Baseline photographs of Mohs surgical wound defects were taken at time of treatment and at additional follow-up visits throughout the protocol. Photographs were analyzed using the open-source program ImageJ,²⁹ and resulting wound area, perimeter, and depth were calculated and recorded. Rate of wound healing was analyzed using standard techniques.^30,31 Specifically, a linear regression using the ordinary least squares method was used to determine the relationship between the wound area and time after the Mohs procedure. The slope of the regression was used as the measure of wound closure, and normalized wound closure rates were used to correct for initial wound size and therefore represent percent wound closure. Statistical significance between the wound closure rates in each treatment group was determined using a one-way analysis of variance (ANOVA) with Tukey post-hoc comparison. The Grubbs test (or extreme studentized deviate test) was used for outlier analysis. Dependence of the rate of wound closure on initial defect size was determined using Pearson product moment correlation. All statistical analyses were performed using Prism (version 9.0; GraphPad).

A total of 25 patients volunteered to participate in the study. These patients were randomly assigned to one of the 2 treatment groups or to the control group. For the purposes of this study, the initial skin cancer diagnosis was not considered. As this study focused on the potential effects of ATDA on the rate of wound closure and cosmetic outcomes, the wound closure properties were deemed to be independent of the diagnosis that resulted in the Mohs surgery. The diagnosis, treatment group, wound location, and original defect size are summarized in Table 1. Listed defect sizes were obtained from patient charts, and quantification of wound area measured from photographs was used in subsequent analyses.

The rate of wound closure was analyzed by measuring the wound area (relative to calibration ruler) at each patient visit for all available wound photographs. As shown in Figure 1A, the rate of wound closure was approximately linear for most cases. The slope of each line determined by linear regression was used as the wound closure rate for subsequent analyses. Analysis of the rate of wound closure showed no statistically significant difference even after the significant outlier in the membrane group was removed from the analysis (Figure 2). 

The data next were analyzed for potential differences in initial defect area as there is a relationship between initial defect size and time to closure. As seen in Figure 3, there were no significant differences between the treatment and control groups with or without the significant outlier in the membrane group. The data were further evaluated by analyzing the correlation between differences in initial defect size and closure rate. As shown in Figure 4, a strong dependence of wound size and rate of wound closure was observed, demonstrating that larger wounds healed at a slower rate compared with smaller wounds (R² = .9069, P < .0001). These findings suggesting a relationship between initial wound size and wound healing rate are consistent with findings of other studies.^30,31

Given the relationship between initial wound size and closure rate, the relationship between wound closure rate normalized for the size of the initial defect was analyzed. As shown in Figure 5A, there was no significant difference between normalized wound healing rate and time to closure (P = .0765 ANOVA; membrane vs control intervention, P = .0802). As shown in Figure 5B, however, the difference between control and membrane closures rates reached statistical significance between control intervention and membrane when outliers were removed (P = .0349 ANOVA). Specifically, with the removal of the control group outlier, the difference between membrane and control intervention was significant (P = .0288), although control compared with liquid treatment was not (P = .2590). 

Because anatomical location affects rate of wound closure, closure of wounds in different anatomical regions also was analyzed. Although there were 3 regions (face, hands, ear) for which there were more than 5 patients with wounds in those regions, comparison between control intervention and treatments with similar sized wounds was possible in only 1 region (ear). For similar sized wounds within about 10% of the initial defect area, both liquid and membrane treatments appeared to increase the initial closure rate when compared with control intervention (Figure 1B and 1C, respectively). Each represents only a single comparison, and any possible treatment effect is small. Because the initial wound size was small, however, a single treatment had little opportunity to increase healing rate.

Because the limitations of the small sample size and uneven distribution between treatment groups did not allow for the detection of significant differences in wound closure, the authors assessed the rate of wound closure and cosmetic outcomes in particularly difficult cases with initial defects of varying sizes and in different locations. Figure 6A documents the healing of 2 patients treated with liquid ATDA after undergoing Mohs micrographic surgery. A patient with a defect measuring 1.6 cm² after removal of a basal cell carcinoma on the right side of their nose was treated with 0.8 mL of liquid ATDA, which was injected around the periphery of the wound at the first postoperative visit. The patient considerered plastic surgery and opting out of the study due to concern for the cosmetic outcome. After treatment, however, the wound completely healed in 33 days with minimal scarring and the patient was entirely satisfied with the outcome (Figure 6A, upper panel). In a second patient, an initial defect measuring 2.3 cm × 2.7 cm that had resulted from removal of a basal cell carcinoma was treated by injecting 1 mL of liquid ATDA at the first postoperative visit after the Mohs procedure. The wound was entirely healed in 22 days without the need for a graft (Figure 6A, lower panel).

Two representative cases in which membrane was used to treat patients post-Mohs micrographic surgery are illustrated in Figure 6B. Membrane was applied to a wound on the temple of a patient at the first postoperative visit. The wound healed in 29 days with no retraction of eyelid or lateral canthus (Figure 6B, upper panel). In a second patient, after removal of a dermatofibrosarcoma protuberans from the second distal toe, membrane was applied at the first postoperative visit. At 42 days after the procedure, the defect was largely filled and the toenail had begun to regrow (Figure 6B, lower panel).

The data presented in this study show that ATDA is effective in healing wounds by secondary intention post-Mohs micrographic surgery and may increase wound healing rates when compared with wounds of similar size and anatomical location treated by standard of care, as seen herein in the cases of the ear. The data do not show robust differences in wound healing rates between ATDA and control intervention with most analyses. When data were normalized to initial defect size, however, a significant difference between membrane and control intervention was observed (P = .0288). This is a notable finding and suggests that the use of ATDA may be beneficial in the postoperative care of routine post-Mohs patients.

A limitation encountered in this study was the variety of wound sites and areas; both the location of the wound and the initial defect size affect the wound closure rate. To more effectively evaluate the efficacy of ATDA in the treatment of wound healing after Mohs micrographic surgery, matched wound location and size will be necessary to provide a direct comparison of wound closure rate. Furthermore, tobacco use (current, recent, or remote) was not assessed and should be included in future studies. Despite the inability to show a robust effect on wound closure rate with ATDA, no adverse events or complications were associated with use of ATDA in these patients. Additionally, all patients treated with ATDA were satisfied with wound closure rates and cosmetic outcome. 

The lack of published articles analyzing comparisons between the use of ATDA vs standard of care for post-Mohs wounds healing by secondary intention, this preliminary case series sought to provide insight into the use of ATDA. Future studies should have the goal of further evaluating ATDA and rate of wound closure post-Mohs micrographic surgery, effectively controlling for the aforementioned limitations. Future investigations of ATDA should also include patients with diabetes as well as healing of post-Mohs wounds, particularly in difficult-to-heal regions such as the lower extremities. 

Use of ATDA in this setting appears to be safe, with patient satisfaction and cosmetically pleasing outcomes. Although patient satisfaction is subjective, patients may consider incurring additional cost for the level of cosmesis achieved and to avoid more invasive closures such as those requiring greater reconstruction involving grafts. Amniotic tissue–derived allograft warrants further investigation because use of such allograft may provide an alternative to wound closure following Mohs surgery in difficult-to-heal locations that may require secondary intention healing or skin grafts.

The authors thank the Mohs surgeons at Affiliated Dermatology who performed the procedures, Midwestern University for funding, and AminoTechnologies for providing the amniotic tissue allografs used in this study. 

Authors: Kayleen Seaton, DO¹; Dustin Mullens, DO²; Jason Barr, DO^1,2; Elizabeth Hull, PhD³; and Richard Averitte, MD^1,2

Affiliations: ¹Arizona College of Osteopathic Medicine, Midwestern University, Glendale, AZ; ²Affiliated Dermatology & Affiliated Laboratories, Scottsdale, AZ; ³Midwestern University, Glendale, AZ

Correspondence: Elizabeth Hull, PhD, Professor, Midwestern University, Biomedical Sciences, 19555 North 59th Avenue, Glendale, Arizona 85308; ehull@midwestern.edu

Disclosure: The authors disclose no financial or other conflicts of interest.