Pyoderma Gangrenosum: Recognition and Management

Pyoderma gangrenosum is an unusual cause of skin ulcerations that wound specialists must be prepared to recognize. There is no diagnostic test since it is a diagnosis of exclusion, and if the disease is not recognized it can quickly become much worse. Pathergy, whereby a lesion begins or worsens due to trauma, such as a scrape, bite, debridement, surgery, or biopsy, is seen with pyoderma and requires special consideration. This case series and review will summarize the salient features of pyoderma and its treatment with an emphasis on the controversial role of surgery with pyoderma.

Brocq first described pyoderma in 1916.¹ In 1930, Brunsting, Goeckerman, and O’Leary coined the term pyoderma at the Mayo Clinic (Rochester, MN) and initially felt it was an infectious etiology.¹

Now, more than 80 years later, the exact etiology is not known, but it is not felt to be infectious and remains a challenge to diagnose and to treat. The name itself adds to the confusion because it is neither gangrenous nor infectious. Very few wound care providers see it on a regular basis and yet many will encounter at least 1 case at some point in their career.

Consideration of the diagnosis is the critical fist step, because if the disease is present but unrecognized, the results can be devastating. In cases of pyoderma, the usual paradigms of wound treatment such as debridement can precipitate pathergy and create a vicious cycle of the wounds increasing in size.

For example, case 1 shows a patient with postoperative pyoderma that occurred behind the ear and had several debridements by an ear, nose, and throat service at a teaching hospital. Even after the wound clearly became worse after each intervention, pyoderma was not recognized as the cause, as seen in case 1 (Figures 1A-1C). 

History. The most common history for patients with pyoderma is of a small lesion thought to be a bite or scratch that rapidly grew in size, or following a medical procedure such as surgery or even something as seemingly innocuous as a needle biopsy, as in case 2 (Figure 2) and case 3 (Figure 3A-3D).

Patient population. Some patients with pyoderma have known inflammatory conditions such as inflammatory bowel disease (IBD) (ie, Crohn’s disease or ulcerative colitis) or other diseases such as lupus, leukemia, or hepatic disease. The percentage of patients with pyoderma who also have IBD was originally quoted at around 50%, but subsequent reviews have seen fewer percentages of pyoderma patients with associated IBD.² Several explanations have been given for this apparent increase of isolated pyoderma over time, such as many of the original series were from tertiary referral hospitals specializing in IBD, or clinicians have become more comfortable over time in diagnosing isolated pyoderma.²

Presentation. Pyoderma lesions may present de novo or following minor trauma, a bite, or surgery.³ Peristomal pyoderma is essentially a postsurgical presentation that has been recognized since 1984.⁴ Peristomal pyoderma is especially common in patients with IBD following surgery and creation of a stoma. While peristomal pyoderma may be more readily recognized because of its location, it may be misdiagnosed from irritation of the appliance or an infection.

Nonperistomal pyoderma can present anywhere on the body, and case 4 (Figures 4A-4C) shows a case of pyoderma presenting on the hands. While there are other generally traumatic or infectious causes of hand wounds, whenever unusual lesions present on the hands—especially if there are multiple lesions—it should trigger immediate consideration of pyoderma. 

Appearance and culture results of lesions. The classic pyoderma lesion is a full-thickness ulcer with bluish purple undermining borders (Figures 4A, 4C). There is no definitive histologic feature, but histologic changes that support the diagnosis are neutrophilic infiltration of the dermis and cutis.

Despite the fact the wounds often have considerable white cell debris that look like pus in the undermining areas, cultures have most often shown no growth in the author’s experience. Positive cultures certainly do not rule out pyoderma; however, pyoderma wounds can become colonized or infected and have positive cultures, especially if they have been open for a prolonged period.

Role of biopsy. Biopsy is used in potential pyoderma cases to rule out other causes for the ulcers. Weenig et al⁵ described a misdiagnosis rate of about 10% in their series from the Mayo Clinic and from literature review.⁵ They concluded biopsy is always warranted because the treatment of pyoderma will often embark a patient on a course of systemic steroids. They classified the nonpyoderma cases that were misdiagnosed as pyoderma into the following groups: vascular or occlusive disease, vasculitis, malignant process, primary infection, drug induced or exogenous tissue injury, and other inflammatory conditions.

Weenig and coauthors⁵ also noted that venous stasis wounds are the most common lower extremity wounds but usually do not look like pyoderma. Venous stasis ulcers generally are not as painful and have associated hemosiderin deposition. Vasculitic ulcers can be difficult to distinguish from pyoderma, and they also often respond to steroids. In addition, they noted that antiphospholipid antibody syndrome is “a highly problematic simulator of pyoderma” due to its appearance and response to steroids, but that the lupus anticoagulant was nearly always positive with antiphospholipid syndrome and ulcers.

Case 5. This case shows a patient with vasculitis due to antiphospholipid antibody syndrome, which can be similar to pyoderma. The presentation of a painful wound is somewhat similar to pyoderma, but the purpura were more widespread (at least during a time of the vasculitis flare up) than typically seen in pyoderma. The prednisone had to be increased when the vasculitis flared up, and an extreme halo of purple was noted. Vaculitis is similar to pyoderma in its response to steroids and also in the potential benefit of using human allograft (Figures 5A-5E). 

Role of biopsy. Clinical judgement makes the diagnosis of pyoderma, not biopsy, so in this author’s practice, clinicians have not felt biopsy is mandatory in diagnosing and treating every case of pyoderma, but that it should certainly be employed liberally. Decisions on when to biopsy are made on a case-by-case basis. Table 1 lists the factors considered when deciding whether or not to biopsy; when more left-column factors are present, the more comfortable avoiding biopsy becomes (Table 1). 

One reason clinicians may be hesitant to biopsy lesions thought to be pyoderma is to avoid the possibility of potentiating pathergy and have the wounds worsen. In the author’s experience, this risk can be avoided by having the patient on systemic steroids starting before the biopsy and continuing until pyoderma is ruled out.

When to hospitalize? Patients with rapidly progressing lesions and/or extensive involvement of the process, or failure to respond to outpatient management, should be considered for hospitalization. Hospitalization is often appropriate to allow close observation, intravenous antibiotics, wound and pain management, and multidisciplinary care.

Critical period of uncertainty. Because pyoderma can only be diagnosed by exclusion, there is a critical period of uncertainty that requires close observation and intense therapy. During this time, the diagnosis of pyoderma will have been chosen based on the clinical presentation, steroids begun, and cultures and/or biopsy obtained. The presence or absence of a systemic disorder will be considered, and appropriate testing is also ordered. As emphasized by Havlik and coauthors,⁶ systemic steroids should be started as soon as the diagnosis of pyoderma is considered likely, and the patient must be closely followed. During the initial treatment, the patient’s response to steroids must be monitored closely while the results of biopsy, cultures, and/or other tests are pending.⁶

This critical period of uncertainty is particularly discomforting for clinicians because atypical infections and necrotizing infections can resemble pyoderma, and the steroids appropriate for initial management of pyoderma may exacerbate these conditions. Local signs that the process is indeed pyoderma responding to steroids would be a decrease in pain, no new lesions, and diminishing halos around the existing lesions within 24-48 hours. If local signs were not improving, an assessment of the overall picture including vital signs and lab results is warranted; if this shows a worsening clinical situation, steroids would be held or quickly tapered.

Dosage of systemic steroids. The author’s treatment protocol has been to start most hospitalized severe potential cases of pyoderma on 100 mg (daily in single dose) of IV methylprednisolone for several days, then taper down to 60-80 mg/day for several days before switching to oral prednisone at 40-60 mg/day and tapering as tolerated. This dosage of steroids is similar to recommendations of other reports.^7,8 The author prefers to start the initial doses intravenously as opposed to orally for hospitalized patients, mainly to speed delivery and avoid any questions of absorption since some of the patients may have IBD. Conversely, for less severe presentations being treated on an outpatient basis, oral administration is appropriate.

A review of the cases and of Table 5 in the Holt⁸ series is particularly instructive: remission most commonly required 60-100 mg/day and doses just under that level were unsuccessful in many cases. Hence it is important to stress that treatment for pyoderma should not be timid—steroid dosage must be adequate (the author suggests 80-100 mg/day) because if there is no good clinical response it is more likely that pyoderma is not present instead of the patient having been undertreated. Regarding the tapering process, in the author’s practice and others it has been noted that tapering by more than 10-20 mg/day every week can lead to flares or recurrence.^2,3

Acute prednisone or methylprednisolone treatment has also been recommended at 1-2 mg/kg/day. For patients of average body weight or less, that would be similar to the regimen noted above.^9,10 Most sources recommend administering the daily amount as 1 dose, but dividing the doses has also been described.⁹ It is not known if adjusting for weight is beneficial, but for patients substantially above or below average body weight it may be prudent to consider dosing based on weight.

Johnson and Lazarus¹¹ described another regimen called pulse dosing, which they defined as the “use of single or multiple daily infusions of suprapharmacologic doses of methylprednisolone.” They presented 3 patients with pyoderma who did not respond to 40 mg or less of prednisone a day but who did respond to 1,000 mg (1g) of IV methylprednisilone for 5 days. In all cases, the pyoderma responded positively with regression of the lesions. Johnson and Lazarus note the pulse therapy is an experimental treatment modality with several deaths reported within 24 hours of pulse treatments. This author’s experience and many others since the Johnson and Lazarus study have established that in most instances methylprednisilone or prednisone at 100-120 mg daily will induce remission of pyoderma.^2,8 Hence, in the author’s practice, clinicians refer to daily doses of IV methylprednisolone in the range of 100mg/day as “mini-pulse” treatment.

Most patients in the author’s practice have eventually been weaned off steroids completely but about half of them have had recurrences that required restarting or increasing steroid doses during tapering for at least a short time period (case 3, Figures 3A-3D). Other series have described a recurrence of about one-third.⁹ For patients who cannot cease steroids without recurrence, steroid-sparing agents can be considered such as cyclosporine, tacrolimus, infliximab, dapsone, and sulfasalazine, among others.^7,9,10,12-16 A good summary table with doses of systemic therapeutic options for pyoderma is found in Callen and Jackson,¹² and Wollina¹⁵ provides a good review as well. Referral to a rheumatologist or other practitioner experienced with using immune suppression therapy may be indicated for patients not easily weaned from steroids.

Local steroid treatment of pyoderma wounds. Small pyoderma wounds that are not progressing rapidly may respond to topical steroids, and the author has used this approach for peristomal presentations with some success.¹⁴ Intralesional steroids have also been described, but the author does not have experience using that technique.⁷

Dressings and surgical care. Gentle dressings and avoidance of undue trauma and irritation are best for pyoderma wounds. Many dressings are documented as having been used, including foams, hydrocolloids, and silver sulfadiazine.¹⁷ Clinicians at the author’s practice have had especially good results with foam dressings containing the aniline dyes methylene blue and crystalline violet (Hydrofera Blue Foam, Hollister Wound Care, Libertyville, IL), and this has been reported in the literature as well.¹⁸ It is possible the aniline dyes suppress inflammation, although this is not known.

Surgery and pyoderma. For extensive wounds with or without necrosis, surgical therapy may be necessary for debridement and/or obtaining coverage. Review of the older literature could easily leave one with the impression that pyoderma absolutely precludes surgery. Tay and colleagues’ “Acute Pyoderma Gangrenosum Does Not Require Surgical Therapy”¹⁹ left no doubt as to how they felt about surgery for pyoderma. They described how progressive debridements of a finger wound worsened the situation, but this was a case of unrecognized pyoderma and no steroid suppression was employed.¹⁹ Similarly, Hickman and Lazarus² stated that due to pathergy, “Grafting attempts are extremely hazardous since grafts rarely take and new ulcers may develop at donor sites. Amputation should be discouraged since pyoderma gangrenosum may arise in the amputation stump.”²

As the author’s experience and the more recent literature makes clear, while the potential for pathergy must always be remembered, it does not, in fact, preclude surgery for debridement or skin grafting if the patient is on steroids.

Numerous examples of surgical procedures succeeding with pyoderma are now known.^{6, 20, 21} Havlik’s 1998 article “Pyoderma Gangrenosum of the Breast: Sequential Grafting” presents the case of pyoderma in the breast of a 15-year-old girl who benefited from initial allografting with skin from a human donor for pain and wound management followed by autografting.⁶ They noted that in some cases autografting will give a much better cosmetic result than healing only by secondary intention. This experience preceded a similar result in case 6 of the current paper (Figure 6). The initial allograft acts not only to control pain and stabilize the wound but, if it is successfully incorporated, it is an indication that autografting will be able to take. Allograft alone may be enough to help close smaller wounds, but because it is typically eventually rejected, autografting remains the most functional and cosmetic option for larger wounds.

The size of the wounds once clean will dictate how to get coverage: relatively small wounds can epithelialize; medium-sized wounds do well with human allograft; and very large wounds may require autografting. Case 6 was an extreme case that demonstrates several important surgical and wound principles. This patient required multiple debridements, which were safely completed without creating pathergy. The resulting large defect had exposed bone and tendon, and allografting was used initially to ascertain whether the wounds would successfully take a graft, thereby avoiding a donor site if indeed the harvested skin would be rejected. Once the allograft was seen to be taking, the patient went back to the operating room for an autograft from the thigh, which eventually closed most of the wound, and there were no problems healing the large donor site.

Recognition of pyoderma is of utmost importance. Hospitalization is helpful for new onset cases with significant wounds. Biopsy will not make the diagnosis but can be used to rule out other causes and, when utilized, should be done under steroid suppression to prevent pathergy when pyoderma is possible. When pyoderma is a possibility, clinicians must be prepared to start steroids and endure the critical period of uncertainty. Surgery may precipitate cases of pyoderma like any other trauma and these presentations may be among the most difficult to recognize compared to infection. Contrary to previous reports, surgery is possible in cases of pyoderma and may be necessary in some cases to improve eventual healing.

From Private Practice

Address correspondence to:
Matthew Q. Pompeo, MD 
healerone@aol.com

Disclosure: The author discloses no financial or other conflicts of interest.