A novel topical therapy for resistant and early peristomal pyoderma gangrenosum

Abstract

Peristomal pyoderma gangrenosum (PPG) is an under‐recognised and difficult condition to treat. We describe a case series using a novel topical combination therapy that promotes wound healing and allows for adhesion of the stoma appliance. A crushed oral prednisolone tablet mixed with Stomahesive Protective Powder (ConvaTec) was applied topically to seven patients with PPG and resulted in pain relief and wound healing in six of seven patients. Only one patient experienced recurrence. The novel topical therapy we describe is cost‐effective, readily available, and easily applied in any inpatient or outpatient setting.

1. INTRODUCTION

Pyoderma gangrenosum (PG) is defined as a neutrophilic dermatosis that presents with painful, necrotic, and undermined ulcerations.1 The incidence of PG is estimated to be 3 to 10 cases per million population annually,1 although these figures were primarily dependent on case series rather than rigorous prevalence and incidence studies. An Italian 12‐month observational, multicentre study reported an incidence of 5.17 new cases per million population per year.1 PG is a debilitating and very painful condition that was first reported in 1916 by Brocq, a French dermatologist,2 whilst peristomal pyoderma gangrenosum (PPG) appears to have been first described by Barker et al in 1975.3 Regardless of anatomical location, PG can be difficult to diagnose and treat.

A definitive diagnosis of PPG is often delayed because of clinical confusion with other peristomal skin ulcerative aetiologies. This can lead to poor treatment decisions that can have adverse consequences on healing outcomes and impacts the patients' quality of life and their ability to undertake activities of daily living.4 Numerous systemic and topical therapies have been trialled with varying success.4

The clinical features associated with PPG include painful and rapid cutaneous ulceration that presents as a circular or arcuate lesion with an undermined red‐violaceous border.5 Wound cultures usually only grow commensals, and skin biopsy histology is often non‐specific.4 Pathergy is a major feature of this neutrophilic dermatosis and generally precludes biopsy and wound debridement.6 The ulceration can progress to expose underlying muscle and fascia.5 The need for stomal appliances to adhere to the skin for the collection of effluent can make management of PPG challenging.

2. LITERATURE REVIEW

PG is an auto‐inflammatory neutrophilic dermatosis involving activation of the innate immune system characterised by infiltration of neutrophils into the dermis and hypodermis associated with ulceration.7 Peristomal PG has been found to be associated with inflammatory bowel disease (IBD), such as Crohn's disease and ulcerative colitis.4 Genetic links such as PAPA (pyogenic arthritis, PG, and acne), PASH (PG, acne, and suppurative hidradenitis), PASS (PG, acne, supparative hidradenitis, and ankylosing spondylitis), PAPASH (pyogenic arthritis, PG, acne, and suppurative hidradenitis), and PsAPASH (psoriatic arthritis, PG, acne, and suppuratives hidradenitis) syndromes are now recognised.1, 8 Furthermore, an analysis of the lesions demonstrates overexpression of the genes producing cytokines, such as IL1‐beta, IL‐17 tumor necrosis factor alpha (TNFα), and IL‐8.5

PG usually starts as a small sterile pustule or nodule and swiftly progresses to having demarcated wound edges with undermined violaceous borders and a surrounding area of erythema.7 Skin tissue is rapidly destroyed, and the red‐blue undermined border is a result of the liquefactive necrosis.9 The base of the ulcer may develop granulation tissue, necrotic tissue, and/or purulent exudate.10 Two clinical presentations of PG have been described, and they are Type R, which is described as explosive and rapidly spreading, and Type G, which is typified as slow to progress and spread.11

The pain caused by PG ulcers can be described as “exquisite” and out of proportion to the ulcer size or depth.7, 9, 12, 13, 14 Cribriform scarring may form as the ulcers heal.7, 9 A diagnosis is one of exclusion and is made on characteristic clinical features, along with the exclusion of all other potential aetiologies.7, 9, 12, 13

Sheldon et al15 conducted a case series of patients with PPG, and they proposed that a biopsy was unnecessary for diagnostic purposes as it failed to provide clinically useful information, although a preponderance of literature suggests that the purpose of a biopsy is also to exclude infection and other differential causes of ulceration.7, 9, 10, 14 Ahronowitz et al7 indicated that a biopsy may be linked to the risk of inducing further pathergy, which is the hyper‐reactivity of the skin in response to minimal trauma.16 If a biopsy is required, an elliptical incisional biopsy has been found to have reduced pathergy compared with a routine punch biopsy.7 Furthermore, it has been hypothesised that trauma caused by the removal of the adhesive ostomy appliances could be a major contributing factor to PPG.16 Although the concept of pathergy associated with PG is highly probable, other impacts on cutaneous perfusion and de‐innervation have not been explored and are an area for future research.

Ahronowitz et al.7 and Su et al.9 recommended the following criteria for PG diagnosis:

• Rapid progression of a painful, necrolytic, cutaneous ulcer with an irregular, violaceous, and undermined border.

• Exclusion of other causes of cutaneous ulceration.

There should also be at least two of the following:

• A history suggesting pathergy or a clinical finding of cribriform scarring,

• Systemic diseases associated with PG, and

• Histopathology findings of sterile dermal neutrophilia with or without mixed inflammation and lymphocytic vasculitis.

Afifi et al.4 reviewed 335 cases of PPG reported in the literature and found that 67% were female, 81% had associated IBD, and 70% demonstrated a flare in their IBD. PPG is reported to be twice as common in patients with Crohn's disease compared with those with ulcerative colitis, and it also has an association with malignancy and rheumatological conditions.4, 17 In this cohort, the mean time to occurrence of PPG following surgery that resulted in stoma construction was 23 months. Most of these patients had undergone surgery that resulted in an ileostomy (78%), whilst patients with a colostomy accounted for 16%, and patients with a urostomy comprised another 6%.4 Interestingly, a study of 259 patients in a German facility with unspecified PG found that 25.5% of patients had diabetes mellitus.17 Therefore, clinicians would be unwise to rule out a diagnosis of PPG when IBD is not an existing comorbidity.

The literature demonstrates that more females than males are affected by PG.6, 10, 15, 18, 19 A retrospective analysis of 259 patients from 20 dermatological wound centres in Germany found a higher prevalence in females than males, with the average age of women being 60 years compared with men at 54 years of age.17 Female gender was also found to be more prolific in a 313‐patient cohort study conducted in the United Kingdom, where 59% were women.20 These findings suggest that females have a higher risk for contracting PG, and the reason for this discrepancy has not yet been determined. A potential rationale for the increased incidence of PG in post‐menopausal women could be the decline in oestrogen levels as it is recognised that oestrogen is beneficial for promoting the deposition of granulation tissue and reducing inflammatory responses.21, 22

Steroid therapy, either oral or topical, is the most often initiated treatment for PG.4 Systemic immunosuppressive treatment agents such as corticosteroids, methotrexate, mycophenolate, azathioprine, and cyclosporine have demonstrated an average complete PG healing response, ranging from 48% to 64% depending on the treatment.4 Oral cyclosporine, dapsone, and metronidazole have been used historically, either alone or in combination with steroids, for the treatment of PG.4 The TNFα antagonists infliximab, eterancept, and adalimumab have a reported complete healing response rate of 50% to 67%.4 Overall, the studies associated with the use of these agents in the treatment of PG are generally comprised of small case series, and this makes for difficulties in determining efficacy and levels of evidence.

Intra‐lesional steroid therapy appears to be less efficacious (34%) and, in some instances, has been proposed to advance the PPG lesion, which may be because of the induction of pathergy as a result of the injection trauma.4, 7 The literature does report some successful outcomes when other topical therapies such as cyclosporine and tacrolimus have been used.4 Surgical closure of the stoma invariably facilitates healing of the PPG, but repositioning of the stoma to another anatomical location often (67%) results in the recurrence of PPG at the new site.4

The literature demonstrated that healing of PPG lesions in patients treated with topical steroids produced a successful healing outcome in 34% of patients.11, 23, 24, 25 A combination of 0.3% of topical tacrolimus with a hydrocolloid and pectin paste (Orabase by ConvaTec), compared with topical 0.05% clobetasol propionate alone, demonstrated an increased healing rate.23 Orabase is a hydrocolloid paste containing pectin and has been used extensively for treating aphthous ulcers.26

Afifi et al.4 also described the use of crushed dapsone and prednisolone tablets and topical application of corticosteroids in either nasal or pulmonary spray formulation with some success, whereas Demartyn et al27 described the successful treatment of PPG using crushed prednisolone tablets combined with a hydrocolloid powder.

3. METHOD

Within our case series, PPG lesions were treated with crushed 5 mg tablets of prednisolone mixed with equal parts of hydrocolloid powder (Stomahesive Protective Powder), and this was applied topically to the peristomal ulcerated areas. The anatomical deficits of deep ulcerated lesions were filled with the prednisolone powder mix, which was then covered with a hydrofibre silver dressing (Aquacel Silver by ConvaTec) followed by a hydrocolloid and alcohol paste, which is commonly used to fill and level peristomal anatomical creases (Stomahesive Paste by ConvaTec) in order to form a level and secure surface for the application of a conventional ostomy appliance. This routine was undertaken on a daily basis. The potential side effects of topical steroid, such as delayed wound healing, exacerbation of local infection, and systemic absorption, were weighed against the therapeutic risk.

Informed consent was obtained for all patients to commence this regime. Written informed consent was also obtained for publication.

4. RESULTS

4.1. Case 1

Mrs A received multiple other treatment interventions for PPG between April 2013 and September 2014. Initially, her therapies involved moderated doses of oral prednisolone (40 mg daily) with no healing benefit. Other ineffectual systemic therapies included: oral minocycline 50 mg daily, dapsone 50 mg daily, mycophenolate 500 mg twice daily, infliximab infusions every 2 weeks, and potassium iodide 300 mg three times daily. Topical therapies applied during this time included numerous intra‐lesional kenacort injections and topical tacrolimus and diprosone. For Mrs A, this entailed 94 outpatient reviews under various medical specialities, including: dermatology, urology, endocrinology, ophthalmology, nephrology, haematology, general surgery, general medicine, and stomal therapy. She required 14 inpatient admissions during this time, totalling 102 bed days, for recurrent urosepsis, pulmonary emboli, rectal bleeding, and dyspnoea, all thought to be associated with steroid‐induced weight gain and proximal myopathy. Her chronic renal failure progressed during this time, and her creatinine increased from 160 mmol/L to 210 mmol/L. The PPG ulcers at their worst were 10 mm × 10 mm × 5 mm deep, 30 mm × 22 mm × 15 mm deep, and 40 mm × 35 mm × 25 mm deep (Figures 1, 2, 3).

Figure 1.

Mrs A before treatment

Figure 2.

Mrs A healed at 1 year

Figure 3.

Mrs A at 3 years with no further treatment or recurrence

An extensive review of the PPG literature at that time indicated a paper written by Demartyn et al,27 which outlined the application of crushed oral prednisolone tablets mixed in equal portions with a hydrocolloid powder (Stomahesive powder) to the PPG lesions. As a rescue therapy, Mrs A was commenced on this treatment in September 2014 in conjunction with 30 mg of oral prednisolone daily. At 1 month, a decrease was noted in slough in the wound bed, and oral prednisolone was rapidly weaned to 4 mg daily in view of her severe steroid side effects. At 3 months, epithelialisation of the wound edges had occurred, and healthy granulation tissue in the wound bed was evident. Mrs A also reported a noticeable decrease in pain. At 1 year, all three PPG lesions had completely healed with no subsequent recurrence. All maintenance therapies were ceased.

4.2. Case 2

Mrs B was a 74‐year‐old female with type 2 diabetes, peripheral arterial disease, peripheral neuropathy, diabetic retinopathy, coronary artery disease, and cerebrovascular disease, who was admitted to hospital with rectal bleeding. Mrs B was subsequently diagnosed with rectal cancer (T4 N0 M0) and underwent an ultra‐low Hartmann's procedure and formation of a colostomy. Postoperatively, she experienced circumferential peristomal dehiscence at the mucocutaneous junction, which required a revision of the stoma 2 weeks later. Severe peristomal skin pain was noted at 1 month following this revision, and she presented with multiple areas of peristomal necrotic eschar. At this time, a punch biopsy was obtained, and the histology report demonstrated ulceration with an underlying mixed dermal inflammation and reactive stromal alteration. These features were non‐specific and could have been related to the presence of a non‐healing, chronic wound. However, these atypical lesions were described as being compatible with PPG. Based on our previous positive experience with case number 1, crushed prednisolone therapy mixed with equal portions of hydrocolloid powder was applied on a daily basis. Significant healing was noted within 10 days, and complete healing was achieved by 4 months (Figures 4, 5, 6).

Figure 4.

Mrs B before treatment

Figure 5.

Mrs B at 10 days post‐commencement of topical therapy

Figure 6.

Mrs B healed at 4 months

A further five patients presenting with PPG were treated with the same treatment regime. The total number of patients treated with this combination therapy was seven, of which six resulted in complete healing. Of the cohort, one patient experienced PPG recurrence, which also healed using the same treatment method. This patient who did not respond to the treatment had her stoma reversed. Results are outlined in Table 1.

Table 1.

Patients treated with prednisolone and stomahesive powder

Age when PPG was diagnosed	Gender	Comorbidities	Time to PG from stoma formation	Number of lesions	Wound size (W × L × D)	Time to pain relief	Time to healing	Diagnostic criteria
62 (Mrs A)	Female	Interstitial cystitis Type 2 DM (poorly controlled) Diabetic peripheral neuropathy Hypertension Early diabetic neuropathy	31 mo	3	① 10 mm × 10 mm × 5 mm ② 30 mm × 22 mm × 15 mm ③ 40 mm × 35 mm × 25 mm	Approx 1 mo	372 d	Biopsy Wound characteristics Pain response
75 (Mrs B)	Female	Coronary artery disease Peripheral arterial disease Type 2 DM Diabetic retinopathy Critical ischaemia (L foot) with hyperbaric therapy Diabetic neuropathy Rectal cancer	33 d	Multiple	Multiple and superficial	15 d	121 d	Biopsy Wound characteristics Pain response
63 (Mr C)	Male	Rectal cancer Type 2 DM (well controlled)	84 d	2	105 mm × 5 mm and 2 mm × 2 mm × 2 mm	21 d	63 d	Wound characteristics Pain response
77 (Mrs D)	Female	Interstitial cystitis	89 mo	1	15 mm × 15 mm × 1 mm Recurrence ① 10 mm × 10 mm × 2 mm ② 10 mm × 10 mm × 2 mm	14 d 7 d	77 d 35 d	Wound characteristics Pain response
66 (Mrs E)	Female	Short gut syndrome	52 d following stoma refashioning	1	15 mm × 52 mm × 1 mm	21 d	113 d	Wound characteristics Pain response
36 (Mrs F)	Female	Ulcerative colitis 1 month post‐partum	53 d	2	① 2 mm × 2 mm ② 15 × 15 × 5mm	Nil	N/A—stoma reversed prior to healing	Wound characteristics Pain response
53 (Mrs G)	Female	Rectal cancer	189 d	1	10 mm × 12 mm × 3 mm	65 d	65 d	Wound characteristics Pain response

Note: Biopsies consistent with PPG.

Abbreviations: DM, diabetes mellitus; PG, pyoderma gangrenosum; PPG, peristomal pyoderma gangrenosum.

5. DISCUSSION

The case series demonstrated that six of seven patients healed with the application of 5 mg of topical prednisolone mixed with equal amounts of Stomahesive powder. Amongst the patients who demonstrated healing, decreased pain was reported in all except one patient within 3 weeks. Although our cohort was small, it was reflective of the literature findings in that three of our eight patients were also type 2 diabetics. Gender discrepancies were noted in our small cohort as six of the seven patients were female. Of the six females, five were post‐menopausal, and the other woman presented in the immediate post‐partum period. In the tertiary hospital, during the period of 1 January 2012 to 31 December 2018, there were 1002 patients who underwent surgery that resulted in stoma formation. Amongst these patients, there were 11 cases of PPG diagnosed, and this accounted for an incidence rate of 1.1%.

We hypothesise that the healing responses to the treatment in our patient cohort was because of the continual steroid bioavailability at the site of inflammation. Stomahesive Protective Powder was used in all of our PPG cases and is a formulation of equal parts of three polymers, sodium carboxymethylcellulose (Na CMC), gelatin, and pectin. All three polymers are biocompatible and biodegradable hydrocolloids and have, either individually or in combination, been investigated previously for their wound‐healing properties.28 The powders form a gel that is similar to hydrogels, which are known to form a barrier that prevents bacterial infection, and their high water content maintains a moist wound bed that promotes a wound‐healing environment.29, 30

Na CMC, gelatin, and pectin have also been widely applied as drug‐delivery carriers, particularly to modify the release of drugs, including prednisolone.31, 32, 33, 34 However, there has been no investigation into the bioavailability of prednisolone presented as a physical mix with the three polymers in combination. Prednisolone is a small‐molecule drug (360.45 g/mol) that is very sparingly soluble in water (223 mg/L). Being a neutral molecule, its dissolution is not pH‐dependent. However, the diameter of the dissolving prednisolone particles, as well as the extent of wetting of the particles, would be critical to the dissolution of this hydrophobic drug. Gelatin kneaded into a physical mixture with prednisolone at a weight ratio of 1:1 has been shown to enhance the dissolution of the drug by improving its wetting in aqueous media, and this has translated to higher initial serum drug concentrations when the kneaded mixture was fed to beagle dogs.35

We hypothesise that the polymers in the hydrocolloid powder, when intimately mixed with the ground prednisolone tablet, would provide a delivery vehicle that facilitates the dissolution and sustained release of the drug in the wound site. Upon contact with and absorption of the wound exudate, the polymers in the powder mix would form a hydrogel matrix. Gelatin, and possibly also pectin and Na CMC, also acts to facilitate wetting and dissolution of the drug particles embedded throughout the hydrogel matrix. The gel‐like matrix modulates the ingression of water and the outward diffusion of dissolved drug to provide a sustained release of the drug into the wound area.

It is possibly helpful that the Stomahesive powder forms a crust at the wound surface. This barrier prevents water evaporation to maintain a moist environment that not only helps with wound healing but allows for continual drug dissolution and bioavailability. Applied neatly for 7 days in a mouse model of skin ulceration, hydrocolloid powder has been reported to stimulate the formation of granulation tissue that is 2.86 times thicker than film dressings.36

Daily reapplication of the prednisolone: hydrocolloid powder mix further ensures a continual drug pool for the wound site. Hydrocolloids are also known to have an autolytic component in the wound‐healing process, thereby facilitating the removal of non‐viable tissue and promoting a healthy wound bed.31 In addition, pectin and Na CMC, being anionic polymers, could regulate the wound pH to provide an acidic environment that is conducive to wound healing.37

Limitations to this case series include that the numbers are small and that all patients used the same hydrocolloid powder; therefore, this may not applicable to other hydrocolloid powders or similar preparations.

6. CONCLUSION

Although this was a small case series, the interventional therapy was relatively cheap, readily available, and easy to use in an outpatient setting under the supervision of experienced clinicians. Furthermore, the topical treatment avoided the necessity for oral or parenteral therapy and associated inherent toxicities.

CONFLICT OF INTEREST

None of the authors have received any funding or support nor are they employed by ConvaTec.

Pearson WA, Prentice DA, Sinclair DL, Lim LY, Carville KJ. A novel topical therapy for resistant and early peristomal pyoderma gangrenosum. Int Wound J. 2019;16:1136–1143. 10.1111/iwj.13164