Pyoderma Gangrenosum

With dermatology, as in all facets of medicine, it is possible to do everything right with regard to assessment and treatment yet still have poor outcomes. These outcomes can be the result of lack of adherence to recommended treatment, an improperly read laboratory or pathology report, unforeseen interactions or reactions to medication or, as in this case study, a rare condition that mimics others in which the comorbidities are the same but the treatments differ greatly.

Reassessing our diagnoses based on outcomes is as important as reassessing our patients for improvements. Continuing a treatment that is ineffective will inevitably lead to poor outcomes, lower patient satisfaction and an increased risk of complications associated with the correct diagnosis.

Presentation

A 90-year-old, highly active and functional man presented to the office complaining of a painful ulceration on his left lower leg (Figure 1). It had been present for more than 2 years. He had been receiving treatment with dressings and frequent debridement for what was diagnosed as a stasis ulcer by professionals at a wound care clinic. The patient reported that the ulcer had not improved with these treatments and in fact it had enlarged and become more painful.

His history included mild hypertension, hypercholesterolemia and type 2 diabetes. He had recently completed treatment for colon cancer diagnosed about 18 months prior. He had been referred by the wound care team for a second opinion and treatment options.

Figure 1. The initial presentation. Figure 2. The ulceration after 1 month of appropriate treatment. Figure 3. The ulceration after 2 months of appropriate treatment. All photos courtesy the author

Objective Findings

This patient had 1+ to 2+ pitting edema on both legs from the upper pretibial regions downward. Pedal pulses were faint but present bilaterally. A 4.5 x 3.5 cm ulcerated, erythematous patch was evident on the left lower leg. The region was tender to the touch and showed signs of maceration. No drainage, radial streaking, increased warmth or other signs of infection were present. The base was clean and pink after recent debridement, and the lesion showed signs of mild undermining of the peripheral margins of otherwise normal appearing skin.

Differential Diagnoses

Stasis ulcer: This ulceration signifies the end stage of severe, untreated stasis dermatitis caused by venous insufficiency of the lower legs. Venous insufficiency is the result of damaged or incompetent valves and blood vessels in the lower extremities. The result is blood pooling in the lower legs that causes edema. If untreated, venous insufficiency can progress to form painful ulcerations known as stasis ulcers. These are most commonly located on the pretibial aspects of the legs but may occur on the ankles as well. These ulcers are well defined with necrotic bases within a field of the stasis dermatitis.¹

Necrobiosis lipoidica diabeticorum: This rare condition typically affects the lower extremities of patients with diabetes mellitus. Despite its association with diabetes, the severity of the diabetes or its comorbidities does not affect the incidence of the disease. In fact, no correlation between its development and the control of blood glucose levels has been observed. The condition usually begins with multiple erythematous papules on the pretibial aspects of the lower legs. As the papules grow, they coalesce to form well-demarcated, atrophic, shiny, yellowish-brown telangiectasic plaques. In most cases, several isolated plaques are present at the same time. If untreated, some of these plaques progress to form painful ulcerations with erythematous borders.²

Pyoderma gangrenosum: This rare but potentially serious ulcerative skin disease is characterized by deep ulcers with a well-demarcated bluish or violet border. The edges of the ulceration appear to undermine the borders, leaving the surrounding tissue erythematous and indurated. The most common locations for these ulcerations are the pretibial regions of the lower extremities. Patients typically describe the associated pain as mild to severe. Development of this disease has been noted in patients with inflammatory bowel disease, gastrointestinal malignancies, rheumatoid arthritis, leukemia and lymphomas.³

Assessment & Plan

A swabbed bacterial culture was collected to rule out infection. The results of this culture showed the growth of normal skin flora and no infectious or bacterial elements. A biopsy was taken, and it included the normal surrounding skin, the undermined margin and the ulceration. The result of the biopsy was inconclusive. No evidence of stasis change was present in the sample. The biopsy result also ruled out necrobiosis lipoidica diabeticorum. The sample showed a significant inflammatory process with neutrophils predominating. This result was suggestive of pyoderma gangrenosum.

The lack of extensive edema or other signs of significant peripheral vascular disease made the diagnosis of a stasis ulcer unlikely. Additionally, the patient had been receiving excellent wound care for 2 years, yet the ulceration worsened. The biopsy results and prebiopsy clinical assessment helped eliminate the initial diagnosis of stasis ulcer. The wound care team was informed that treatments should be discontinued, since debridement might be worsening the condition.

This history of disease progression was atypical for necrobiosis lipoidica ulcers. The ulceration had not started as erythematous or pruritic papules. The patient reported that the ulceration formed "like a sink hole." The only evidence that something was forming was mild discomfort and redness at the site. The ulceration began shortly after these symptoms surfaced, and it had been increasing in depth and diameter since. The biopsy results and lack of additional lesions or plaques also supported the exclusion of necrobiosis lipoidica as a potential diagnosis.

The inconclusive biopsy results, patient history and, most importantly, the clinical picture and progression of the ulceration, led to the diagnosis of pyoderma gangrenosum. Pyoderma gangrenosum is a noninfectious neutrophilic dermatosis. The hallmark signs associated with this disease are recurrent painful ulcerations with undermined borders and violaceous borders. Although it is possible for lesions to develop elsewhere, the most common site is on the lower legs, as was the case with this patient.⁴

The patient was seen at a clinical stage in which the ulcer was larger and deeper than what is typical. In most cases, pyoderma gangrenosum lesions begin as a follicular pustule that develops rapidly, causing tissue necrosis and an advancing ulceration.⁴ The initial pattern of ulceration has a "cat's paw" appearance prior to necrosis, at which time the lesion becomes more uniform.³

The patient in this case had a history of colon cancer - and it turns out a long history of ulcerative colitis that was not initially disclosed. Ulcerative colitis is present in 10% to 15% of patients affected by pyoderma gangrenosum. Other diseases that have been associated with its development include Crohn disease (10% to 15% of cases), hepatitis C, seronegative rheumatoid arthritis, ankylosing spondylitis, lymphoma and leukemia.4 Although absent in this case, pathergy, the development of ulcers at the site of trauma, is common.³ This tendency likely resulted in worsening of the condition as a result of frequent debridement of the wound. The lack of a bluish-to-violet margin can also be attributed to the wound care the patient had been receiving.

The epidemiology of pyoderma gangrenosum is not well understood or documented. The peak onset is between ages 20 and 50, with women affected slightly more often than men. Pyoderma gangrenosum has been documented in patients older than 65, but the occurrence rate is low. The condition's association with inflammatory bowel diseases has led to the hypothesis that pyoderma gangrenosum may be caused by a cross reaction of the antigens produced in the bowels, exerting an effect on the skin.⁴

Early diagnosis of pyoderma gangrenosum is challenging. It is, in essence, a diagnosis of exclusion. The history is pivotal, since it may be the only thing that gives a clue to the underlying cause of the ulcerations. There are no laboratory parameters, and histology is nonspecific. In the early stages, histopathology shows only deep suppurative folliculitis with dense neutrophilic infiltrate.⁴ At the time the biopsy was performed in this case, the ulcer had been present so long that follicles were not identifiable and only the neutrophilic infiltrate was present.

Treatment

Several treatment options are available for pyoderma gangrenosum, all of which produce inconsistent results. Treatments focus on suppressing the neutrophilic immune responses in the skin. Systemic therapies are commonly employed because they can slow progression of the ulceration and minimize the duration of the disease. High-dose prednisone 1 mg/kg/day to 2 mg/kg/day is commonly prescribed, but due to potentially fatal side effects, its use should be avoided in patients with a history of cardiovascular disease and diabetes.⁴

Another systemic medication approach to treating pyoderma gangrenosum is cyclosporine A in combination with prednisone. Studies have documented good results when these are used in combination; cyclosporine A does not produce sustained results when dosed as monotherapy.

A combination of sulfa drugs dosed with dapsone has become a popular treatment choice. The dapsone inhibits neutrophil migration and acts as an anti-inflammatory. Immunosuppressants such as azathioprine and tacrolimus have been used with varying effectiveness; results are inconsistent. Anti-tumor necrosis factor alpha medications such as etanercept and infliximab have been used in combination with oral prednisone with mild success.³

Topical therapies have limited effectiveness in the treatment of this disease. High-dose topical steroids delivered under occlusion have demonstrated some success, as has topical tacrolimus. These have limited capabilities and are used for mild cases.³ Tacrolimus has greater effectiveness in the variant of pyoderma gangrenosum associated with peristomal wounds.⁴

In the described case, the patient expressed a desire to avoid oral medications due to digestive problems that developed after his treatments for colon cancer. Additionally, he and his wife were concerned about the effects of oral prednisone on his diabetes. The fragility of his skin caused by age and months of debridement and dressings made high-potency steroids under occlusion problematic. With this in mind, I prescribed topical dapsone gel 5% applied twice daily. I instructed the patient to wear a dressing only when necessary, to avoid further maceration.

The topical form of dapsone was selected because of the success associated with its oral variant. Although indicated for inflammatory acne, I hoped that the antineutrophilic and anti-inflammatory effects of this medication would induce healing of this ulcer. The patient returned to the office every 2 weeks for 2 months as the ulceration healed. At 1 month, the diameter of the ulcer had decreased by 50% and the depth was appreciably less (Figure 2). At 2 months, no ulceration was noted and re-epithelialization was apparent (Figure 3). At 3 months, only postinflammatory pigmentation was noted. The patient continues to improve.

Discussion

When evaluating cutaneous ulcerations, it is important to avoid becoming fixated on a single diagnosis. In this case, the appearance and location of the ulcer supported the initial diagnosis of stasis ulcer. The lack of response to the prescribed therapy should have prompted a reevaluation of the diagnosis. Unfortunately in this case, it took 2 years for this to happen. If the response to therapy is not what is expected, reassess and change the diagnosis and therapeutic approach as indicated.

References

1. Wolff K, et al. Chronic venous insufficiency. In: Fitzpatrick's Color Atlas and Synopsis of Clinical Dermatology. 5^th ed. New York : McGraw-Hill; 2005: 475-479.

2. Scaramuzza A, et al. Case study: Necrobiosis lipoidica diabeticorum. Case Rep Pediatr. 2012;2012:152602.

3. Brooklyn T, et al. Diagnosis and treatment of pyoderma gangrenosum. BMJ. 2006;333(7560):181-184.

4. Wollina U. Pyoderma gangrenosum-a review. Orphanet J Rare Dis. 2007;2:19-26.

Raymond Shultstad is vice president of the National Academy of Dermatology Nurse Practitioners. He practices at the Center for Dermatology and Skin Surgery in Spring Hill, Fla. He has completed a disclosure statement and reports no relationships related to this article.

Click to view larger graphic.