When Epithelialization Beats Granulation in Sacrococcygeal Stage IV Pressure Ulcers/Injuries: A Report of Two Cases

© 2024 HMP Global. All Rights Reserved.

Any views and opinions expressed are those of the author(s) and/or participants and do not necessarily reflect the views, policy, or position of Wounds or HMP Global, their employees, and affiliates.

Wound healing is the process by which the body repairs and regenerates damaged tissue. It involves complex interactions between many cell types and signals and occurs in 4 sequential stages: hemostasis, inflammation, proliferation, and remodeling.

An effective hemostasis stage minimizes blood loss from the wound. In response to injury, surrounding blood vessels vasoconstrict and a platelet plug forms to seal areas in the blood vessel wall that are damaged and bleeding. Simultaneous activation of the coagulation cascade results in the formation of fibrin, which transforms the platelet plug into a stable, crosslinked clot. During this stage, activated platelets release growth factors, cytokines, and vasoactive molecules from their cytoplasm, initiating the inflammatory stage.¹

The goal of the inflammatory stage is to clear debris and prevent infection. Blood vessels dilate to allow immune cells, signaling molecules, enzymes, antibodies, and other nutrients into the wound bed. Neutrophils arrive within 24 to 36 hours to phagocytose bacteria, foreign particles, and damaged tissue, and then undergo apoptosis and form slough.¹ Within 48 to 72 hours, macrophages arrive and begin to clear the apoptosed neutrophils. Macrophages also release growth factors and signaling molecules to activate keratinocytes, fibroblasts, and endothelial cells, triggering the proliferative stage.^1,2

During the proliferative stage the wound bed is filled with new granulation tissue, infiltrated by blood vessels, and resurfaced via epithelialization.³ Fibroblasts are the dominant cell of the proliferative stage. They produce MMPs, which degrade the fibrin clot, as well as ECM components such as type III collagen, fibronectin, proteoglycans, and hyaluronan, which form a provisional matrix.³ Angiogenic signals prompt new blood vessels to form and grow inward from the wound periphery to support the developing tissue.^1,2 Other signals trigger wound edge keratinocytes and stem cells from hair follicles and sweat glands to migrate, proliferate, and form an epithelial cover over the wound. These processes are regulated by MMPs, growth factors, cytokines, integrins, keratins, chemokines, and extracellular macromolecules.³ The granulation tissue formed during the proliferative stage provides a scaffold for future cell adhesion, migration, growth, and differentiation. The blood vessels provide nourishment, and the epithelial cover keeps the underlying tissue protected while it undergoes further modification.

Modifications to the organization and composition of the granulation tissue matrix are made during the remodeling stage. Type III collagen laid down during the proliferative stage is replaced by type I collagen.⁴ Myofibroblasts contract the wound to decrease its surface area. Angiogenesis stops, and blood flow and metabolic activity of the wound return to preinjury levels.⁵ After wound healing is complete, the resultant tissue has up to 80% of the tensile strength of uninjured skin.² The actual tensile strength achieved depends on multiple factors, including patient age, health, and nutritional status.⁶

Successful progression through the stages of wound healing requires a complex interplay of cellular and molecular signaling and the correct physiological environment. Under favorable conditions, most wounds exhibit a reduction in surface area of at least 15% per week, resulting in approximately 50% healing within 1 month.⁷ Wounds that exhibit delayed reduction in surface area or arrested progress through the expected stages of healing are classified as chronic wounds.⁸

Chronic wounds occur when the cellular and molecular environment are not favorable for progression through the stages of wound healing. Often, stagnation occurs in the inflammatory stage. Factors such as hypoxia, ischemia, infection, excessive inflammation, altered cellular response, defects in collagen synthesis, reperfusion injury, edema, and pressure can impede wound healing.⁶ These factors can result from systemic and local disease states such as diabetes, chronic kidney disease, cirrhosis, and osteomyelitis, as well as obesity, advanced age, immobilization, impaired circulation, malnutrition, smoking, and alcohol consumption.^6,8 Medications that interfere with wound healing include cytotoxic antineoplastic and immunosuppressive agents, angiogenesis inhibitors, corticosteroids, nonsteroidal anti-inflammatory drugs, and anticonvulsants.⁹ Patients with multiple medical comorbidities can have multiple risk factors simultaneously and therefore are at increased risk of developing chronic wounds.

Chronic inflammation in chronic wounds leads to an impaired transition into the proliferative stage, and therefore to impaired granulation, angiogenesis, and epithelialization. Chronic wounds have relatively more matrix degradation via MMPs than matrix protection via TIMPs.^10,11 As a result of this imbalance, granulation tissue is degraded. Chronic wounds also have a surplus of inflammatory cytokines such as tumor necrosis factor α, and relative lack of proliferative cytokines such as platelet-derived growth factor and transforming growth factor β, resulting in decreased mitotic activity.¹⁰ Fibroblasts in chronic wounds are unable to make collagen at a rate fast enough to overcome degradation. In addition, some fibroblasts in chronic wounds demonstrate premature senescence.

Chronic wounds also undergo altered epithelialization. The keratinocytes at the nonhealing edges in chronic wounds are phenotypically and biologically different from edge keratinocytes in acute wounds. The nonhealing edge keratinocytes are mitotically active in both the basal and suprabasal layers, whereas healing edge keratinocytes are mitotically active in only the basal layer.¹² Chronic wound keratinocytes are also more likely to exhibit parakeratosis and hyperkeratosis.¹² Epidermal growth factor receptors are internalized in keratinocytes of chronic wounds, so they are unable to respond to extracellular signals.¹² In addition, the imbalance of MMPs and TIMPs seen in chronic wounds impairs keratinocyte migration. As a result of these differences, chronic wounds are more likely than acute wounds to have a nonmigratory, hyperproliferative epidermis that does not reepithelialize.

The current report discusses 2 patients with multiple medical comorbidities who had full-thickness, chronic wounds that did not progress through the stages of wound healing in the order that theory dictates. Instead, their wounds underwent epithelialization without granulation. Consequently, new skin grew inward and covered the wound before the proliferative stage built up the wound bed. The resulting epithelial surface of the healed wound was deeper than the surrounding tissue. Remarkably, these patients achieved wound closure without fully experiencing all the customary stages of wound healing.

Case 1

A 51-year-old bedridden female with a history of paraplegia due to multiple sclerosis was referred to the Toledo Hospital Wound Center, ProMedica Health Network in Toledo, Ohio for wound management. She presented with a stage IV pressure injury (Figure 1A), and MRI confirmed the diagnosis of osteomyelitis. On physical examination, the bone could not be palpated, and interventional radiology staff were consulted and a CT-guided bone biopsy performed. Culture results from bone biopsy grew Escherichia coli and methicillin-susceptible Staphylococcus aureus. The patient’s osteomyelitis was treated with intravenous antibiotics for 6 weeks per the recommendation of an infectious diseases specialist. However, due to stool contamination of the wound, the patient underwent a diverting colostomy.

After the wound was clean, the patient was treated for 12 months with collagen dressing and intermittent negative pressure wound therapy when the wound became macerated. Although the wound area decreased, wound depth did not significantly improve. Granulation tissue did not proliferate at the base of the wound to obliterate the defect; instead, epithelium began to advance from the periphery of the wound. In addition, epithelial islands progressively became visible within the wound bed.

The progression of the wound healing was protected with a negative pressure wound therapy sponge and cover layer dressing made of knitted cellulose acetate fabric and impregnated with a petrolatum emulsion (Adaptic; Solventum). With continued offloading of the ulcer using a low air loss mattress and frequent repositioning along with nutritional support with a high protein diet and multivitamin supplement, the wound bed became bowl-shaped, with approximately 70% epithelialization (Figure 1B).

Two applications of fish skin xenograft were attempted as ECM replacement over a period of 1 month, but granulation tissue still did not fill in and instead, complete epithelial coverage took place (Figure 1C, D). As of this writing, the patient continues to follow up and the chronic osteomyelitis remains stable, with low biomarkers of C-reactive protein (<2 mg/dL) and erythrocyte sedimentation rate less than 20 mm/hour.

Case 2

An 86-year-old bedridden male with a myriad of medical comorbidities, including kidney stones with hydronephrosis, urinary tract infections, and a long history of slow healing stage IV sacrococcygeal pressure injury (Figure 2A), underwent treatment at the wound clinic for 2 years during healing. He also had chronic osteomyelitis treated previously by infectious disease with follow-up CT and MRI demonstrating stability.

Initially, the wounds were treated with wet-to-moist dressing changes using antiseptic solution (diluted 0.125% strength Dakin solution) and then switched to daily calcium alginate silver dressing changes after cleansing with soap and water. During the treatments, it was noted that robust granulation tissue was not present. Instead, peripheral epithelium had advanced to cover the entire wound bed. In some areas, hyperkeratotic epithelium formed, which was then selectively debrided (Figure 2B, C). The wound was kept clean, with daily cleansing and applications of neutral petroleum jelly.

The wound eventually healed over a 6-month period with use of a low air loss mattress, and offloading (Roho cushion; Permobil) with frequent repositioning, and a high-protein diet with multivitamin supplements.

At the 1-year follow-up visit, the wound remained covered with a healthy bowl-shaped epithelium. The wound was cleansed every other day with soap and water, and petroleum jelly was applied daily. The wound appeared to have contracted, resulting in a smaller and shallower final defect (Figure 2D).

The healing process of full-thickness wounds has been well-studied, occurring in the 4 stages previously discussed: hemostasis, inflammation, proliferation, and remodeling. Full-thickness wounds involve injury to the epidermis, dermis, and subcutaneous tissue. Deeper structures, such as bone, can also be involved. Partial-thickness wounds involve only the epidermis and dermis. Because fewer layers of tissue are involved, partial-thickness wounds can heal by means of reepithelialization without the requirement of initial granulation tissue formation, allowing for a faster healing process.¹³

Bornes et al¹³ hypothesized that partial- and full-thickness wounds use different cellular and molecular reepithelialization strategies. In full-thickness wounds, keratinocytes migrate to cover the entire wound bed while simultaneously the entire epidermis coordinates wound closure and reestablishment of epidermal homeostasis by migration, proliferation, and differentiation. Three models of keratinocyte migration and reepithelialization in full-thickness wounds have been proposed: the leap-frog model, the collectively migrating multilayer model, and the sliding model.¹³

In the leap-frog model, suprabasal keratinocytes slide over leading basal keratinocytes, roll into the wound bed, connect with the basement membrane, and migrate to cover the wound.¹³ In the collectively migrating multilayer model, both suprabasal and basal keratinocytes migrate together, forming an epidermal sheet that closes the wound. In the sliding model, the basal keratinocytes at the leading edge migrate across the wound bed, bringing the physically connected suprabasal keratinocytes with them and forming an epithelial sheet that covers the wound. Bornes et al¹³ conclude that a fourth model is observed in partial-thickness wounds, in which basal keratinocytes migrate in a collective sheet across the wound, and suprabasal keratinocytes stay in place. This reepithelialization strategy allows keratinocytes to avoid possible immobile obstacles on the skin (eg, hair follicles), allowing for faster formation of a protective skin barrier.¹³

The cases described in the current report demonstrate reepithelialization occurring before (or in the absence of) oblierating granulation tissue. Granulation was likely absent because the patients’ medical comorbidities created environments that were not conducive to wound healing. Both patients had decreased mobility with pressure injuries and treated chronic osteomyelitis, resulting in wound beds that were poorly perfused, frequently infected, and chronically inflamed.⁶ As mentioned in a previous section, ongoing inflammation in chronic wounds impairs the transition from the inflammatory stage to the proliferative stage, and therefore impairs granulation, angiogenesis, and epithelialization.³ The imbalance of MMPs and TIMPs and the surplus of inflammatory cytokines in chronic wounds leads to increased degradation and decreased production of granulation tissue.^10,11 Epithelialization is also impaired in chronic wounds because of phenotypical and biological differences in the edge keratinocytes that result in a nonmigratory, hyperproliferative epidermis that does not cover the wound bed when attempting to achieve healing via the full-thickness models.¹²

It is possible that the comorbidities and chronic inflammation in the patients in the current case report prevented granulation tissue formation and epithelialization via the full-thickness models. However, epithelialization was still achieved via the partial-thickness model proposed by Bornes et al¹³ because that model does not require prior granulation.

The limitations of this case report include its design as a case series with a small number of cases because of the rarity of this occurrence. Additionally, as of this writing there are no studies demonstrating reepithelialization of full-thickness wounds without initial granulation. Discussion about the possibility of this process is based on research involving the healing of partial-thickness wounds in vivo in mice.

The current case reports highlight a rare occurrence in which stage IV pressure injuries with considerable depth underwent epithelialization before enough granulation tissue developed to obliterate the defect. The resulting tissue healed with residual depth. The insights gained from these cases suggest the existence of an alternative pathway for epithelialization, possibly resembling the pathway proposed by Bornes et al¹³ for partial-thickness wounds.

Further research is needed to understand the underlying mechanisms and identify factors that contribute to the unique healing pattern observed in the 2 cases discussed herein. Additionally, larger-scale studies involving diverse wound types and patient populations would help validate these findings across multiple settings. If the existence of this pathway proposed by Bornes et al¹³ is confirmed, it would have potential implications for chronic full-thickness wound management because treatments could be developed to specifically target the pathway that would result in accelerated wound epithelialization. 

Authors: Richard Simman, MD^1,2; Monik Gupta, BA³; Anderson Lee, BA³; and Caroline Howell, MD³

Affiliations: ¹The University of Toledo, College of Medicine and Life Sciences, Division of Plastic and Reconstructive Surgery, Toledo, OH; ²Jobst Vascular Institute, ProMedica Health Network, Toledo, OH; ³The University of Toledo, College of Medicine and Life Sciences, Toledo, OH

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

Patient Consent: Consent for the publication of all patient photographs and medical information was obtained. The patients understand that the information will be made publicly available.

Correspondence: Richard Simman, MD; Jobst Vascular Institute, 2109 Hughes Drive, CJT Suite 400, Toledo, OH 43606; Richard.simmanmd@promedica.org

Manuscript Accepted: June 14, 2024