Treatment of Stage 4 Pressure Injuries With Autologous Heterogenous Skin Construct: A Single-Center Retrospective Study

Pressure injuries (PIs) are a challenging problem in health care affecting 2.5 million people per year in the US alone.^1,2 As the population ages and the rates of obesity, diabetes, and cardiovascular diseases increase, a growing number of people will experience decreased mobility and compromised health.^3,4 Consequently, the incidence of PI is expected to increase over the next decade. Notwithstanding significant advances in prevention and treatment, PIs continue to be a prevalent debilitating disease that places a tremendous burden on the affected individuals, health care systems, and society.⁵ Padula et al estimated that the cost of hospital-acquired pressure injuries could exceed $26.8 billion and that 59% of these costs were disproportionately attributed to stage 3 and 4 pressure injuries.²

PIs are classified according to wound depth.⁶ In stage 1 PIs, the skin remains intact but shows signs of persistent inflammation. In stage 2 PIs, the epidermal barrier has been compromised. Stages 1 and 2 are typically associated with moisture and shear but can indicate potential for deeper injury. Stage 3 PIs are full-thickness injuries that involve subcutaneous tissue, whereas stage 4 PIs extend into deep tissues and can involve fascia, muscle, tendon, ligament, cartilage, periosteum, and bone.⁶ Increased wound depth and degree of exposed underlying structures are correlated with greater disease severity and poorer outcomes. The high metabolic demand and length of time necessary to heal large defects is further complicated by an elevated risk of serious infections such as osteomyelitis and sepsis.^1,7-14 One-third of stage 4 PIs progress to osteomyelitis, which significantly increases morbidity and mortality.¹⁵ Patients with stage 3 and 4 PIs have a 5-fold higher mortality rate than all other hospitalized patients, and in the US alone, 60,000 deaths each year are directly attributed to PIs.^2,10,16 

While stage 1 and 2 PIs are often managed with use of appropriate dressings and pressure redistribution support surfaces, stage 3 and 4 PIs are more challenging. Nonsurgical standards of care for stage 4 PI have reported closure rates as low as 5% and 30% at 8 and 12 weeks, respectively,^17,18 and from 9.8% to 30.6% at 6 months.^19-21 Because of this, extensive surgical debridement and reconstruction are often required.^6,22 Early surgical closure of pressure injuries can decrease hospitalization and morbidity rates.^23,24 However, donor site morbidity can be significant, and if the wound recurs or another develops, options for further reconstruction become limited owing to the decreasing availability of donor tissue. Flap dehiscence or failure can leave the patient with a defect much larger than the original wound. Surgical flap reconstruction for pressure injuries has been reported to have complication rates of 25% within 30 days and 59% to 73% overall, with recurrence and dehiscence rates of 29% and 31%, respectively.^25,26 Even with the most advanced treatments and skilled care, PIs often persist for months or years. Clearly, there is a dire need for less invasive and more effective treatment options. 

Autologous heterogenous skin construct (AHSC) is a novel autograft manufactured from a small section of healthy full-thickness skin. Tissue processing results in multicellular segments that retain all endogenous regenerative and supportive cell populations necessary for native wound repair. The investigators sought to determine whether AHSC could provide wound volume reduction, area reduction, and ultimately, wound closure in recalcitrant stage 4 pressure injuries while avoiding morbidity (hospitalization, operations, and adverse reactions). The long-term safety and durability of AHSC was also monitored. 

Institutional review board (IRB) approval (IRB 1731663-2) was obtained for this retrospective cohort study, which met the waiver criteria as described in the 45 Code of Federal Regulations 164.512 (i) (2) (ii). All patients treated with AHSC (SkinTE; PolarityTE) for PIs from January 2020 to May 2021 at a single center were included. The only exclusion criterion was the absence of at least 2 months of follow-up data. There were no restrictions on age, PI etiology, location, duration, or prior treatment. Patient demographics, wound characteristics, and ASHC treatment details of all patients were retrospectively collected directly from the electronic health record. 

The primary efficacy outcome was complete wound closure, defined as 100% re-epithelialization of the wound surface with no discernable exudate, drainage, or need for a dressing. Secondary efficacy outcomes included percent area reduction, percent volume reduction, and coverage of exposed structures. Analysis included donor-site and wound complications, the need for reoperation, and wound recurrence. Data on the number of wound-related hospitalizations, days spent in the hospital, and operations were collected from the first appearance of the wound. The number of events was divided by the data collection time frame for before and after treatment, respectively, to give a paired result for each patient. Rates of events before and after treatment were then analyzed using a Wilcoxon signed-rank test. Additionally, AHSC dosing information was gathered to assess correlations with outcomes. The effect of dosing on the incidence of adverse events was analyzed using bivariate correlations for adverse events coded 0 or 1 for no or yes, respectively.

Treatment Description

All wounds were surgically debrided to viable intact tissue at time of tissue harvest and at AHSC application. Any known infection was treated systemically before tissue harvest. Negative pressure wound therapy (NPWT) was applied to open wounds between debridement procedures. Donor tissue harvesting and AHSC deployment were performed under aseptic conditions. Donor tissue was obtained from an area of healthy skin, typically the trunk or abdomen. A small, full-thickness ellipse of the skin, including a layer of hypodermis, was excised using a sharp surgical scalpel. Hemostasis was ensured, and the donor sites were closed primarily under minimal tension.

The donor tissue was placed in a conical tube with saline and shipped overnight at 4°C to a US Food and Drug Administration–registered biomedical manufacturing facility (PolarityTE, Salt Lake City, UT) where it was processed into AHSC. This material was returned (within 2 to 4 days, depending on shipping constraints) in a needleless syringe in the consistency of a paste. AHSC is spreadable and able to cover an area much larger than the size of the donor tissue. Donor tissue is not cultured or grown ex-vivo. 

Each patient’s individualized AHSC was dispensed and spread evenly over all surfaces of their prepared wound bed, including undermined and tunneled areas. A nonabsorbent silicone sheet, fenestrated to allow for fluid egress, was placed in the central open portion of the wound, held against the wound surface with the negative pressure foam, and secured directly to the outer wound margins with staples or sutures. Hydrocolloid wafer was cut into strips and placed around the wound margin to protect the peri-wound from moisture and help maintain the airtight seal of the NPWT dressing (Figure 1). NPWT was left in place for 1 week, and the types of subsequent dressings were determined based on ongoing wound assessment.

Postoperative protocols varied depending on wound characteristics. However, in most cases, NPWT was continued and changed weekly until the wound volume was restored. In all cases, a moist wound-healing environment was maintained with appropriate dressings until wound closure. Protection of the wound from pressure and other mechanical forces was stressed throughout the treatment period. Pressure redistributing wheelchair cushions and mattresses were utilized, and patients were encouraged to limit the amount of time spent in wound-related weight-bearing positions (eg, time spent sitting in a wheelchair for ischial/sacral wounds). Nutritional status, bacterial burden, and all other wound-related factors were monitored on an ongoing basis.

Patient Demographics

Patient demographics are shown in Table 1. Seventeen patients (22 wounds) were treated with AHSC. Of the patients included, 12 (71%) had 1 PI and 5 (29%) had 2 PIs. All treated wounds were stage 4. The mean patient age was 57 years (range, 29-78 years); 6 patients (35%) were female and 11 (65%) were male; 2 patients (12%) were black, 2 (12%) were Hispanic, and 13 (76%) were non-Hispanic White. All patients had a history of infection (drug-resistant infection, 65%; osteomyelitis, 18%; sepsis, 18%), malnutrition, and hospital admission during treatment. More than half (59%) of the patients had diabetes. One patient had a history of single below-knee amputation, and 1 patient had recent bilateral below-knee amputation. Most patients had paralysis (82%), 70% were nonambulatory, all used some form of assistive device for locomotion, and 47% required skilled nursing facility placement. The mean number of wounds experienced by a patient during the treatment period (of any type) per patient was 3.4 (range, 1-11). This included nontreated wounds present at the time of treatment as well as wounds that occurred before closure of the treated wound(s).

Wound Characteristics

The wound characteristics are presented in Table 2. Notably, the mean wound age for the cohort was 29.7 months (±28.4); 59% (13) of the wounds had failed at least 1 attempt at flap reconstruction (range, 1-4), and 95% (n = 21) had exposed bone in their index ulcer at the onset of treatment. The mean initial wound area and volume were 29.4 (SD ± 15.6) cm² and 106.1 (SD ± 112.8) cm³, respectively. 

Outcomes 

Outcomes are shown in Table 3. Complete closure was achieved in 50% of patients in a mean time of 146 (SD ± 93) days. Mean percent area and volume reductions were 69% and 81%, respectively, with a mean follow-up duration of 39 (SD ± 16) weeks. A 95% volume reduction was achieved in 68.2% of patients in a mean time of 106 (SD ± 83) days, and critical structures such as bone and tendon were fully covered in 95% of patients in a mean time of 33 (±19) days. Both are highlighted in the Kaplan-Meier graph in Figure 2. Progression of AHSC in 1 patient with a 14-year-old ischial PI is shown in Figure 3.

Wound-related morbidity results are shown in Table 4. Hospitalizations were found to have a mean annualized decrease of 1.65 per year (P = .001), days spent in the hospital were found to have a mean decrease of 20.92 days per year (P < .001), and number of operative visits was also found to be significant with a mean decrease of 2.36 per year (P < .001).

No donor-site complications were observed. Six patients elected to undergo a repeat application when the initial rate of wound closure began to slow. Two patients experienced recurrence after closure, one closed again within a week, and one persisted in a patient with tetraplegia and inadequate caregiver support. Three serious adverse events (AEs) were reported, all unrelated to the product. One patient had an episode of respiratory failure related to pulmonary infection, one became infected with COVID-19 and died, and 1 patient experienced a pulmonary embolism. One event not included in serious AEs involved a patient with a history of progressive weakness and weight loss that preceded his PI. His wound closed 83 days after AHSC application; however, his weight loss and weakness progressed. He was later diagnosed with cancer and subsequently died. Six AEs unrelated to the index wound were noted (urinary tract infection [n = 2], fracture, suprapubic catheter malfunction [n = 2], and gastrointestinal bleeding). There were 4 AEs related to the index ulcer in 3 patients. One patient was readmitted twice: first for flap closure due to slow healing and again after the flap failed when he was readmitted for a second application of AHSC (445 days after the initial application). A second patient was diagnosed with recurrent osteomyelitis 271 days after AHSC application, and the third patient was readmitted for flap closure 302 days after AHSC application. 

No observable trends were noted with respect to dosing and outcomes. The effect of dosing on the incidence of AEs was analyzed using bivariate correlations for AEs coded 0 or 1 for no or yes, respectively. However, no such relationship was observed. The patients who received the lowest dose (128 µL/cm²) and highest dose (1700 µL/cm²) both healed and had no wound-related adverse reactions.

Medical Complexity

The overall health status of the cohort before treatment was significantly impaired. Variables shown to be significantly predictive of poor PI healing include larger wound size, older wound age, deeper wound depth, presence and number of concurrent wounds of any etiology, presence of infection, older patient age, nonambulatory mobility status, history of dialysis or renal transplant, presence of paralysis, malnutrition, need for long-term or skilled nursing facility (SNF) stay, and/or patient hospitalization for any reason.²⁷ In this cohort, the mean time that treated wounds were present despite appropriate care and advanced wound treatments was 29.6 months. Wounds were both large and deep, were stage 4 (95% with exposed bone), and had a mean wound area and volume of 29.4 cm² and 106.1 cm³, respectively. Approximately 59% of the wounds were in an ischial location, which has also been correlated with poor PI outcomes.^28-30 Collectively, patients in the group had a mean 7.8 comorbidities each, and 76% had additional (range, 1-10) concurrent wounds. All patients showed evidence of malnutrition and required at least 1 hospital stay. All patients in this group experienced mobility difficulties, with 71% being either wheelchair- or bed-bound and 47% requiring SNF placement. 

Impact of Spinal Cord Injury

In this group, all but 3 patients had some level of spinal cord injury, which complicates healing in several ways. Associated muscle wasting depletes protein stores within the body and decreases the amount of protective cushioning over bony prominences. Damage to the neurological system not only causes loss of sensation, making patient compliance with pressure relief measures less likely, but also leads to such issues as recurrent urinary tract and pulmonary infections, which impact overall health. Spasticity can make offloading pressure points more difficult and is a risk factor for developing PI.³¹ Furthermore, lack of innervation interferes with normal wound healing. Neurotransmitters and neuropeptides secreted by cutaneous nerves play a role in all phases of wound healing, including keratinocyte proliferation and re-epithelialization, granulation tissue formation, angiogenesis, wound contraction, and collagen deposition and remodeling.³² Skin denervation has been shown to inhibit skin wound flap healing³³, which may be a possible factor in the failure of 22 surgical flap closures in this cohort. 

Wound Closure

Based on accepted standards, the wounds studied in this cohort would be expected to have a poor probability of healing. This reality, while providing sobering context as to the medical complexity of the group, also makes the observed outcomes more remarkable. 

The reported closure rates for stage 4 PIs range from 9.8% to 30.6% at 6 months.^19-21 Considering the number of variables indicative of poor healing present among this group and the duration of wound persistence despite appropriate advanced wound care management, this group should have healing rates at the low end of reported outcomes. Surprisingly, AHSC achieved closure in 50% of stage 4 PIs in this group, well above the high end of reported outcomes. 

Coverage of Structures and Wound Volume Reduction

Stage 4 PIs often present with large volume defects. In such cases, closure by secondary intention can take months or years to achieve. When bone is exposed, the risk of infection, morbidity, and mortality increases, and restoration of wound volume becomes even more important. Advanced scaffold dressings and NPWT have been used with varying success, but the process is prolonged and requires significant resources. Flap coverage is considered the treatment of choice because of its ability to fill large defects and achieve rapid closure. However, as discussed, complication rates and donor-site morbidity are high, significant hospital resources and highly skilled providers are required, and future reconstructive options are reduced. In the AHSC study group, structure coverage was achieved in a mean time of 30.7 days and wound volume was restored in 68% of patients in a mean time of 15 weeks. While AHSC took longer to restore volume than a successful flap reconstruction might, it was less invasive, presented decreased risk of serious complications, and preserved future surgical options. 

Impact of Mechanical Trauma

Wounds that were better protected from mechanical forces, such as pressure and shear, seemed to heal much more rapidly and be more likely to heal. While this may be intuitive, the difference in closure rates between wounds that were protected and those where this was more difficult (due to lack of caregiver support, patient nonadherence, or other situational issues) was striking. An anecdotal example of this is the progression of 1 patient, a 69-year-old man with C5 quadriplegia who was admitted with a necrotizing soft tissue infection related to a long-standing ischial PI and sepsis. The treated area included the PI and the defect left by surgical debridement of infectious and necrotic tissue and had a volume of 144 cm³. The patient was sent to an SNF after AHSC treatment, which allowed his needs to be met without putting pressure on his wound. The wound was completely closed at his follow-up visit on day 86. 

In contrast, a 64-year-old man with L1 paraplegia and a sacral wound with a volume of 58.5 cm³ had excellent caregiver support. However, on his own admission, he continued to spend most of his time sitting in his wheelchair. While the exposed bone in his wound was covered by day 35 and remained so, on day 393 the wound volume had increased by 12%. Interestingly, although the patient did not adhere to pressure relief recommendations, epithelial islands still formed within the wound bed, signifying continued proliferation and differentiation of regenerative cells within the wound (Figure 4). Similar observations in other patients suggest that AHSC is resilient but significantly hindered by mechanical trauma. Protection from mechanical forces is an important factor to consider with AHSC treatment.

Morbidity and Mortality

Morbidity and mortality in patients with stage 4 PIs is significant. In a review of US hospital discharge records from 9.6 million patients, 43.9% of patients with stage 4 PIs were readmitted within 180 days. Worse yet, 21% had died, representing a mortality rate that was 5 times higher than that of patients with no PI.³⁴ It has been said that PIs, rather than being a causative factor, are often a manifestation of an overall advanced disease state.¹² Remarkably, in this cohort, overall morbidity significantly decreased after AHSC treatment. Patients had a mean decrease of 1.6 hospital admissions, 29.92 hospital stay days, and 2.36 operative procedures per year, suggesting a change in the trajectory of overall disease progression. This was an unexpected finding, and it is unclear why AHSC had such a dramatic effect on health status. One explanation may be that chronic wounds represent an elevated inflammatory state and high metabolic cost for the patient. It is possible that if AHSC helps reverse this by facilitating the transition of wounds from a chronic, inflammatory state to an actively healing, proliferative state, it may have a beneficial effect on overall health. 

AHSC Technology

Chronic wound environments are characterized by the disruption of complex and interdependent physiological processes and signaling pathways necessary for healing. Inflammation, hypoxia, and bacterial burden, among other factors, result in reduced levels of growth factors, elevated protease levels, and impaired cellular activity.³⁵ Owing to its unique characteristics, AHSC has been shown to survive well in difficult chronic wound environments.^36-39 The manufacture of AHSC generates multicellular aggregates that retain all endogenous skin regenerative and supportive cell populations. Processing improves the surface-area-to-volume ratio, which improves cell survival and activates multicellular aggregates that utilize endogenous wound repair support pathways to facilitate wound healing.^40-42 In this study, AHSC performed well in various hostile environments and appeared to rapidly improve the vascularity of the wound bed and shift wounds into a healthier proliferative state.

This was a small retrospective study with limited long-term follow-up. Recurrence rates may increase with longer follow-up in this cohort, but the short-term recurrence rate (2 wounds) is already an improvement over those found in the literature for flap reconstruction. Because this was a single-center study, and because there is lack of standardization in wound care, results do not account for the great variability seen in wound care practice. This study represents wound care done with an experienced interdisciplinary team, including a skilled wound care surgeon, where optimal wound bed preparation and supportive care were employed. Without this, the utility of AHSC would likely be impacted. Patients served as their own controls before and after treatment, and while this method revealed interesting findings, prospective studies with matched controls would add further detail. Multiple modalities were used (NPWT with AHSC), so it is difficult to quantify exactly how much of a role each played in the outcomes; however, it is worth noting that, in most cases, multiple advanced treatments, including NPWT, had been utilized over long periods of time (a mean of over 29 months) without success. The patients treated in this group represent a population that, despite the best efforts of skilled providers and significant allocation of resources, has wounds refractory to healing. Patients in this category tend to accumulate with providers and facilities willing to treat them due to the lack of effective treatment options. Although this study was limited in size and scope, the findings may represent a promising option for the treatment of complex cutaneous wounds that are refractory to standard of care. Additional larger prospective controlled studies are warranted to confirm these preliminary results.

In this retrospective case series, AHSC demonstrated the ability to cover exposed structures and achieve wound volume restoration, area reduction, and durable wound closure in chronic refractory stage 4 pressure injuries. The statistically significant decreases in hospital admissions, days spent in the hospital, and operative procedures noted with AHSC treatment are suggestive of an improvement in overall patient health and may have benefit in reducing the significant socioeconomic burden associated with PIs. AHSC represents a minimally invasive alternative to reconstructive flap surgery that preserves future reconstructive options while minimizing donor-site morbidity and promoting improved patient health.

The authors would like to thank Walter Wills, BSN, for assistance with statistical analysis. During the study period, AHSC technology was registered with the US Food and Drug Administration (FDA) as a human cell, tissue, and cellular and tissue-based product (HCT/P), regulated solely under Section 361 of the Public Health Service Act. It is now being studied as an investigational new drug for the treatment of chronic cutaneous ulcers and is currently available for investigational use only.

Affiliations: ¹Department of Surgery,Lakeland Regional Health, Lakeland, FL; ²Medical Affairs, PolarityTE, Salt Lake City, UT

Correspondence: Diana Burgueño-Vega, MD; diana.burgueno-vega@mylrh.org

Ethics: Institutional review board approval (1731663-2) was obtained for this retrospective study, which meets the waiver criteria, as described in the 45 Code of Federal Regulations 164.512 (i) (2) (ii). 

Data access statement: Research data supporting this publication are available upon request of the corresponding author.

Funding: No funding or grants were received in support of this study. 

Disclosures: MB and RM are employees of PolarityTE. DB and DS have no conflicts of interest to declare.