SAWC Spring 2019 Poster Abstracts

Practice Innovations

Rishabh Gupta, PhD; Marcon Laforet, MSc; Yunghan Au, PhD, MBA; and Sheila Wang, MD, PhD

Introduction. Accurately forecasting wound closure trajectories is difficult due to the incomplete knowledge of wound dynamics. However, the ability to predict wound trajectories would greatly benefit wound care practitioners in identifying high-risk wounds while also determining treatment efficacy.

Objective. Here, the authors demonstrate that models based on big data and modern predictive analytical techniques can adapt to patient-specific variations and be more suited for inference in a clinical setting.

Methods. A predictive model was created to more accurately predict time-to-heal than an existing model based on PUSH scores. The model to predict the trajectory of wound area progression is based on an exponential function using linear mixed-effects model (LMM) methodology. These LMMs incorporate intersubject variances and adapt their predictions to individual subjects/patients. As such, LMMs model the dependent variable as a combination of weighted fixed (population) effects and random (subject-specific) effects.

Results. The proposed model predicted wound trajectory with a goodness-of-fit (r²) value of 0.40. Also, the median error in predicting time-to-heal for wounds with sizes > 1 cm², 5 cm², and 8 cm² for the PUSH-based model was 30, 50, and 55 days, whereas for the proposed model it was 20, 17, and 23 days, respectively. Therefore, the proposed LMM-based exponential model is much more robust in predicting time-to-heal than the previously published PUSH-based model.

Conclusions. The application of predictive algorithms on wound trajectories can be applied to digital wound care management to see if wounds are healing as expected or, alternatively, if treatment strategy needs to be changed. Future work would focus on incorporating more data and more clinically relevant features to improve model performance.

Interested in seeing more posters exhibited at SAWC Spring 2019? Register for the conference and attend the Poster Gala on Thursday, May 9!

Laboratory Research

Amina El Ayadi, PhD; Anesh Prasai, PhD; Jayson Jay, PhD; Ye Wang, BSc; David Herndon, MD; and Celeste Finnerty, PhD

Introduction. Hypertrophic scarring is the most debilitating factor that impedes the quality of life of burn survivors. Though cellular senescence was reported in burn patients’ skin, the temporal activation of cellular senescence in scar development and maintenance is unexplored. Epigenetic modifications and cellular senescence can occur prematurely following major stress (eg, burns).

Objective. To determine the role of senescence activation in the development and maintenance of hypertrophic scars post burn, the authors analyzed the temporal expression of the main senescence-associated secretory proteins (SASP) during scar development using the red Duroc pig (RDP) burn model.

Methods. The RDP burn model closely mimics human hypertrophic scarring post burn. Red Duroc skin biopsies were collected at 24 hours, and 1, 2, and 4 months post burn. The expression of messenger ribonucleic acid (mRNA) transcripts for SASPs and epigenetic modulators were compared with nonburned control biopsies.

Results. The expression of mRNA transcripts of various SASP components were upregulated at 24 hours post burn, including the cell cycle proteins p21 (P < .01) and p16 (P < .001) and the inflammatory cytokines and chemokines IL-1β (P < .01), IL-6 (P < .01), IL-8 (P < .001), MCP-1 (P < .001), TGF-β (P < .01), CXCL-10 (P < .01), and TGF-β (P < .01). The expression of mRNA transcripts for other markers of cellular senescence involved in cell motility and actin polymerization also was upregulated 24 hours post burn (CFL-1, P < .05; PAI-1 , P < .01) before returning to basal levels at 1-month post injury. Loss of lamin B1 expression, characteristic of early senescence activation, also was observed in the model 24 hours post burn (P < .01). Matrix metalloproteinase mRNA expression was upregulated early post burn and stayed elevated for up to 4 months post burn compared with nonburn controls.

Conclusions. These data indicate early activation of SASP contributes to the development and maintenance of hypertrophic scars post burn.

Health Economics

Brian Samsell, BS; Shawn Cazzell, DPM, FAPWCA, FAPWH; and Peter Moyer, DPM, FACFAS, AAPWCA

Introduction. Diabetic foot ulcers (DFUs) with exposed bone or tendon can be difficult to heal. These recalcitrant wounds can create an economic burden because they often require multiple treatment applications. New research is needed to identify cost-effective solutions. Both Dermacell acellular dermal matrix (D-ADM) and viable cryopreserved human placental membrane (vCHPM) have shown success in treating bone-exposed ulcers.

Objective. This analysis compared the cost of each skin substitute using Centers for Medicare & Medicaid Services cost schedules.

Methods. Diabetic foot ulcer healing rates and the number of applications needed to achieve healing were obtained retrospectively from 2 similar studies. The cost of healing 1 DFU in a hospital outpatient department (HOPD) was calculated using [(# of Applications)*(Bundled HOPD Payment + Physician Payment)]. The cost of healing 1 DFU in a physician’s office was calculated using [(# of Applications)*(Product Cost per Application + Physician Payment)]. Treatment cost for HOPDs were extracted from Quarter 3 2018 CMS Addendum B OPPS Payment using CPT code 15275. Skin substitute cost was calculated using October 2018 Medicare Average Selling Price for vCHPM and invoice cost for D-ADM.  The smallest-sized graft for the mean area, 5 cm x 5 cm, was used for calculating cost.

Results. The ulcer areas were comparable, with 13.7 cm² and 14.6 cm² for D-ADM (n = 22) and vCHPM (n = 27), respectively. At 16 weeks, D-ADM had a healing rate of 59.1% with a mean of 1 application needed to achieve healing. The vCHPM had a healing rate of 59.3% with a mean of 9 applications needed for healing. The average cost of healing a single DFU in a HOPD was $1667 for D-ADM compared with $15 007 for vCHPM. The mean cost of healing a single DFU in a physician’s office was $1811 for D-ADM compared with $31 129 for vCHPM.

Conclusions. Despite the equivalent healing rates, vCHPM cost an average of 9 to 17 times more to heal complex DFUs when compared with D-ADM.

Evidence-Based Practice

Marie Brown-Etris, RN, CWON, CCHP; Catherine Milne, APRN, MSN, ANP/ACNS-BC, CWOCN-AP; and Stephen Davis, BS

Introduction. In June 2016, the Wound Ostomy Continence Nurses Society and the Canadian Association for Enterostomal Therapy held a joint conference in Montreal, Canada. The National Pressure Ulcer Advisory Panel’s (NPUAP) revised staging system was presented at this conference. Stage 2 continued to be defined as a partial-thickness loss of skin with exposed dermis, but now included in this revised definition is the statement that granulation tissue is not present. Hence, according to the NPUAP, if granulation tissue is present, then this pressure ulcer is to be categorized as stage 3.

Objective. The purpose of this presentation is to clarify an error that exists in the NPUAP’s revised definition of a stage 2 pressure ulcer. 

Methods. A literature search produced numerous scientific publications and presentations to support the fact that stage 2 pressure ulcers do indeed form granulation tissue.

Results/Conclusions. At SAWC Fall 2018, a laboratory research poster discussed granulation tissue in deep partial-thickness wounds, what would be a considered as a stage 2 pressure ulcer. Based on the NPUAP revised definition, a stage 2 pressure ulcer would be categorized erroneously as a stage 3 pressure ulcer. This has significant and far-reaching implications for clinicians, health care organizations, and the interpretation of research findings, as well as legal, financial, and regulatory implications. For example, it will result in erroneous reporting nosocomial stage 2 pressure ulcers as stage 3 pressure ulcers to the state Department of Health, which may have implications on Magnet status, and there will be an impact on Medicare payments to facilities as well as monetary penalties.  

The anatomy and physiology of skin and the biology of wound repair will be discussed with dermatopathology slides presented to support these facts. The authors’ intent is to provide the audience with the evidence-based facts to support evidenced-based practice.

Interested in seeing more posters exhibited at SAWC Spring 2019? Register for the conference and attend the Poster Gala on Thursday, May 9!

Case Series

Luis Fernandez, MD, FACS, FASAS, FCCP, FCCM, FICS; and Marc Matthews, MD, MS, FACS

Introduction. Negative pressure wound therapy with instillation and dwell time (NPWTi-d) has been utilized in both acute and chronic wounds with positive clinical benefits. A reticulated open cell foam dressing with through holes (ROCF-CC) has been developed to assist with wound cleansing by removing thick wound exudate and infectious materials. This ROCF-CC dressing can provide wound cleansing when debridement is not possible or appropriate in patients. Recently, the use of NPWTi-d with ROCF-CC dressings has been reported with positive outcomes in complex patients.

Objective. The authors report their experience using NPWTi-d with ROCF-CC dressings in 19 patients with complex wounds. 

Methods. Eight of the 19 patients underwent sharp debridement. Oral and/or intravenous antibiotic treatment was initiated in patients as needed. All patients received NPWTi-d with ROCF-CC with instillation of quarter-strength Dakin’s solution, hypochlorous solution, or saline with a dwell time of 5 to 10 minutes followed by 2 to 3.5 hours of continuous negative pressure at -125 mm Hg. Dressing changes occurred every 2 to 3 days.

Results. Ten men and 9 women (average age, 58.2 ± 15.1 years) were treated. Common patient comorbidities included hypertension, diabetes, obesity, and paraplegia. Wound types included pressure injuries, traumatic wounds, and surgical wounds. The average length of NPWTi-d use was 9.5 ± 4.1 days. In all patients, wound beds showed the development of healthy granulation tissue following NPWTi-d. All patients were discharged from care to another hospital facility, skilled nursing facility, long-term acute care facility, or home.

Conclusions. In the authors’ clinical practice, NPWTi-d with ROCF-CC provided effective and rapid removal of the thick exudate and infectious materials and promoted excellent development of underlying granulation tissue.

Clinical Research

Christian Bianchi, MD; William Tettelbach, MD; Niki Istwan, RN; Brandon Hubbs, MS; Kimberly Kot; Stan Harris; and Don Fetterolf, MD

Introduction. Statistical interpretation of data collected in a randomized controlled trial (RCT) is conducted on the intention-to-treat (ITT) and/or the per-protocol (PP) study populations. An ITT analysis is a comparison of treatment groups including all patients as originally allocated after randomization regardless if treatment was initiated or completed. A PP analysis is a comparison of treatment groups including only those patients who completed the treatment as originally allocated, though this is often criticized due to its potential to instill bias. A previous report from an RCT conducted to evaluate the efficacy of dehydrated human amnion/chorion membrane allograft (dHACM) as an adjunct to standard care consisting of moist dressings and multilayer compression in the healing of venous leg ulcers (VLUs) only reported PP study results (n = 109; 52 dHACM, 57 standard care patients) though there were 128 patients randomized: 64 to the dHACM group and 64 to the standard care group. The primary study outcome was the incidence of healing at 12 weeks.

Objective. The purpose of the present study is to report ITT results on all 128 randomized patients and assess if both ITT and PP data analysis arrive at the same conclusion of efficacy for dHACM as a treatment for VLUs.

Results. Healing rates for the ITT and PP populations were 50% and 60% for those receiving dHACM and 31% and 35%, respectively, for those in standard care. Within both ITT and PP analyses, these differences were statistically significant (P =.0473, ITT; P = .0128, PP). Kaplan-Meier plot of time-to-heal within 12 weeks for the ITT and PP populations demonstrated a superior wound healing trajectory for dHACM compared with VLUs treated with standard care alone. 

Conclusions. These data provide clinicians and health policy makers an additional level of assurance regarding the effectiveness of dHACM.

Case Study/Series

Todd W. Greenwood, PhD, MPH; and Matthew J. Regulski, DPM

Introduction. Remote ischemic conditioning (RIC) activates the body’s natural protective physiology against reperfusion injury and the tissue damage caused by low oxygen levels. More than 10 000 patients have completed trials with RIC over the last 2 decades and 20 000 more are currently enrolled. Although much research has focused on prevention and postacute treatment in cardiovascular injury, few trials have focused on the patients with diabetes and diabetic ulcers.

Objective. This case series is early evidence from a proof-of-concept study that aims to show that RIC, delivered with an automated device, is an effective treatment modality for healing diabetic leg and foot ulcers (DFUs).

Methods. Three patients with venous leg ulcers (VLUs) and DFUs were given 40-minute treatment sessions of RIC at their home, 3 times per week, for about a 12-week period (total of 36 treatments). In addition, all received weekly standard of care. Remote ischemic conditioning treatment was delivered with an automated pressure cuff that inflates on the bicep to occlude blood flow for 4 successive 5-minute periods.

Results. Wounds were reviewed on a weekly basis. In this initial series, all ulcers were 100% epithelialized by the end of the treatment period. The VLU (which had been continually treated for 14 months) closed after 10 weeks. The DFUs were healed after an average of 9.3 weeks. Patients reported improvement in neuropathic symptoms and claimed increased energy.

Conclusions. Remote ischemic conditioning treatment was well tolerated by all. Through this series (and the continuing trial), the authors show RIC may be an effective, noninvasive, novel adjunctive approach to treating patients with diabetes.

Oral Abstract

Interested in hearing oral abstract presentations? Register for SAWC Spring and attend the Friday morning session!

Practice Innovations

Ozgur Guler, PhD; Patrick Cheng, MS, MBA; Emmanuel Wilson, MS; and Kyle L. Wu, MD, MBA

Introduction. Accurate wound healing prediction can help guide treatment decisions by early identification of patients presenting with nonhealing risk. Prior efforts include the development of Wound Healing Index for diabetic foot ulcers. Such prediction and risk stratification systems demonstrate some efficacy, but they exclude vital wound image information and associated visual details. 

Objective. Using a data set from a chronic wound data repository, the authors apply the latest developments in deep learning to develop a healing prediction model integrating wound images.

Methods. The proposed deep learning framework combines wound images and patient demographic data to develop a healing prediction model utilizing a Long Short-Term Memory (LSTM) network. An open-source Python deep learning library was employed to develop the machine-learning framework. The model, a combination of Deep Convolutional Network and Deep Neural Network, was trained using the Adam optimizer and categorical cross entropy as the loss function. The model output is the predicted healing trajectory.

Results. The authors tested the proposed model on an input data source comprised of 4 weeks of longitudinal data from the chronic wound data repository. The predicted and actual healing trajectories are compared by way of the Pompeiu-Hausdorff (P-H) distance, which is a measure of trajectory similarity (where zero would denote 2 identical trajectories). The model tested here achieves a P-H distance of 1.6 cm².

Conclusions. The proposed novel LSTM network enables the use of wound image and patient demographic data to perform robust sequence learning. This early application of artificial intelligence may enable more robust healing prediction, thereby facilitating better risk stratification and, subsequently, treatment decisions. Future work will focus on improving the robustness and generalizability of the hybrid LSTM cell with more images and data augmentation.

Oral Abstract

Clinical Research

Elliot T. Walters, MD; Kareem Termanini, MS; Brandon T. Jackson, MS; Tara M. Chadab, MS; Jessica S. Wang, MD; Sarah R. Sher, MD; and Karen K. Evans, MD

Introduction. Over the past 3 decades, advances in transplant surgery have allowed life-saving renewal of solid organ function. Many of these patients are predisposed to the development of lower extremity wounds. There is a paucity of data describing treatment, utilization, and outcomes of this complex patient population. 

Objective. This study examines the surgical outcomes of these patients and overall limb salvage rates in a tertiary care center with a multidisciplinary team focused on limb salvage.

Methods. Medical records were reviewed for all patients who presented with a foot wound and functioning solid organ transplant who received surgery (including debridement or toe/foot/limb amputation) at the authors’ institution from 2010 to 2017.

Results. Fifty-nine patients were identified: 40 with a kidney transplant, 13 with combined kidney-pancreas transplant, and 6 with a liver transplant. The average solid organ transplant patient received 6.5 (± 4.8) procedures during the follow-up period (46 months; range, 6.8-149.4). Thirty-two patients (54.2%) required toe amputations; 15 of 32 (46.9%) of these healed, 14 of 32 (43.8%) progressed to a midfoot amputation, and 8 of 32 (25%) progressed to a below-knee amputation (BKA). Of patients with hind foot wounds, 11 of 15 (73.3%) required calcanectomy; of these, 7 of 11 (63.6%) progressed to a BKA. Overall, 19 of the 59 (32.2%) patients required BKA, including the 15 who progressed from lesser procedures; 2 of the 19 patients requiring BKA progressed to an above-knee amputation.

Conclusions. This retrospective study highlights the morbidity associated with foot ulceration in the immunosuppressed patient population with an overall BKA rate of 32.2%. Solid organ transplant patients who develop lower extremity wounds require a high level of surgical care. A large proportion of these patients will eventually require a major lower extremity amputation. Further research is needed to develop more effective strategies to salvage the lower extremity in the immunosuppressed patient.

Oral Abstract

Clinical Research

Elliot T. Walters, MD; Mira Pandya, DPM; Neha Rajpal, BA; Michel Abboud, BS; Karen K. Evans, MD; John S. Steinberg, DPM; Christopher E. Attinger, MD; and Paul J. Kim, DPM

Introduction. Split-thickness skin grafting (STSG) is an effective coverage method for chronic foot ulcers. Many of these ulcers develop on the plantar aspect of the foot. Split-thickness skin grafting may be a high-risk operation in such a high wear area. However, outcome data are lacking for patients receiving STSG to the plantar aspect of the foot, especially in terms of recurrence. 

Objective. The purpose of this study is to retrospectively compare outcomes of STSG to plantar and nonplantar chronic ulcers.

Methods. Medical records for all patients receiving STSG from 2014 to 2016 at the authors’ institution were reviewed under IRB approval. Major outcome measures included time to healing and time to ulcer recurrence.

Results. There were 158 patients with 182 STSGs identified. Fifty-two STSGs were located on the plantar surface of the foot and 130 STSGs on a nonplantar surface. Percent take at 30 days was not significant between the 2 groups (56% plantar vs. 60% nonplantar, P = .199). However, plantar surface STSGs were significantly less likely to be healed at 60, 90, and 365 days (P < .05). Split-thickness skin grafting on the plantar surface were 3.7 times more likely to experience a grafting complication (OR, 3.7; 95% CI, 1.82-7.50). After 12 months of follow-up, there was no significant difference in ulcer recurrence (17% plantar vs. 10% nonplantar, P = .172).

Conclusions. Split-thickness skin grafting to the plantar surface of the foot is more likely to experience complications and take longer to heal on nonplantar surfaces, but the durability of the graft approximates that of nonplantar surface grafts. This suggests STSG to the plantar surface of the foot may require more upfront investment from the patient and the surgeon but remains a viable wound coverage option.

Oral Abstract

Interested in hearing oral abstract presentations? Register for SAWC Spring and attend the Friday morning session!

Clinical Research

Pheobe Hammer; Emile Latour, MS; Mary Clare Bohnett; Fatima McKenzie; Veselina Korcheva, MD; Stephanie Mengden, MD; Kevin White, MD; and Alex G. Ortega-Loayza, MD

Introduction. Histopathologic evaluation of cutaneous ulcers is indicated when the clinical diagnosis is not well established or when ulcers have not responded to standard of care. Many non-malignant skin ulcers lack specific histologic findings on biopsy and pose a diagnostic challenge. While the usefulness of skin biopsies to diagnose underlying malignancy in ulcerated lesions has been demonstrated in previous studies, their utility in the diagnosis of ulcers of other etiologies has not been reported.

Objective. This study evaluates the role of skin biopsies in the definitive diagnosis of non-malignant ulcers and assesses the level of interobserver variability in the histological diagnosis.

Methods. The study was conducted with random histologic specimens of 45 patients between 2015 and 2018 that included skin biopsies of non-malignant ulcers such as vascular ulcers, autoimmune/inflammatory ulcers, and infectious and dermatitis artefacta. The slides were independently reviewed by 3 dermatopathologists who were blinded to clinical information. These diagnoses were then compared with an independent clincopathologic consensus diagnosis.

Results. The leading diagnosis from each of the 3 raters agreed with the final clinical diagnosis, on average, for 29.6% of the cases (average pairwise κ = 0.15). When asked to provide up to 3 differential diagnoses, any of the raters’ diagnoses and the final clinical diagnosis agreed on average for 25.2% of compared diagnoses (average pairwise κ = 0.10).

Conclusions. Skin biopsies alone are useful in the evaluation of non-malignant ulcers but not sufficient to establish a definitive diagnosis. The low concordance in the histopathologic diagnosis among dermatopathologists also reflects the real-world challenges when assessing these ulcers. Additional workup and clinicopathologic correlation is necessary in the assessment of non-malignant ulcers. Future research endeavors should explore alternative approaches to more efficiently diagnose non-malignant ulcers.

Oral Abstract