Extract
Pyoderma gangrenosum is an uncommon inflammatory skin disease causing progressive, painful cutaneous ulceration. Systemic disorders are associated with approximately 50% of the cases; the most common disorders are inflammatory bowel disease, rheumatoid arthritis and the seronegative arthritides, hematological malignancies, and monoclonal gammopathies, most typically IgA.
Pyoderma gangrenosum is an uncommon inflammatory skin disease causing progressive, painful cutaneous ulceration. Systemic disorders are associated with approximately 50% of the cases; the most common disorders are inflammatory bowel disease, rheumatoid arthritis and the seronegative arthritides, hematological malignancies, and monoclonal gammopathies, most typically IgA.
The bullous variant is often associated with myeloproliferative diseases1,2. Histologically, pyoderma gangrenosum is classified among the neutrophilic dermatoses, a group of cutaneous disorders characterized by a widespread, noninfectious, dermal neutrophilic inflammatory infiltrate.
The histologic findings are variable and depend on the age of the lesion and the location of the biopsy. Although the histopathologic features are not diagnostic, a skin biopsy is necessary to rule out other causes of acute skin ulceration. The location of the biopsy is crucial in order to distinguish the lesion from other conditions, including the other neutrophilic dermatoses, infections, and vasculitis. The most fruitful yield comes from the violaceous edge of the lesion, rather than the center of the ulcer, which may show nonspecific secondary changes. A deep wedge biopsy including both the edge and a more central portion of the ulcer is most useful, and special stains for bacteria, mycobacteria, and fungi should be performed. Because the histologic features of pyoderma gangrenosum are nonspecific, the diagnosis is truly one of exclusion and is heavily reliant on clinical features and course. The absence of diagnostic laboratory tests or pathognomonic features and the fact that associated conditions may be undeclared put the onus on the treating physician to exclude other causes of skin ulceration and to identify any underlying disease. In one recent study, the frequency of misdiagnosis of pyoderma gangrenosum was 10%3.
The condition most often presents on the legs as a rapidly expanding solitary ulcer with a violaceous, undermined border and a necrotic ulcer base. Other characteristic features that are highly suggestive are pathergy (a condition in which the application of a stimulus leaves the organism susceptible to stimuli of a different kind), with new lesions arising at the sites of minor trauma, and cribriform (sieve-like) scarring of healed lesions. The presence of an associated disorder such as inflammatory bowel disease supports the diagnosis; however, an idiopathic form is well recognized and thus clinicopathological correlation is essential in order to arrive at the correct diagnosis. The differential diagnosis includes necrotizing soft-tissue infections such as necrotizing cellulitis and necrotizing fasciitis, deep mycoses, mycobacterial infections, tertiary syphilis, and chronic ulcerative herpes simplex. Ulcerative lesions of pyoderma gangrenosum should therefore be evaluated for a possible infectious cause, and biopsy and appropriate cultures are essential. Other diagnoses to be considered include vasculitis (especially Wegener granulomatosis), vascular insufficiency, and warfarin or heparin necrosis.
The treatment of pyoderma gangrenosum is directed at stabilization of the ulceration, usually with high-dose corticosteroid therapy and local wound care. While there is a role for judicious surgical intervention, premature surgery must be avoided for a number of reasons. Pathergy is a common occurrence, and 15% to 40% of patients have development of persistent ulceration at the sites of quite minor trauma4. Surgical procedures such as débridement and skin-grafting are not recommended during the acute stages, especially in patients who exhibit pathergy, because such procedures may lead to further tissue destruction and progression. Surgery can incite the disease process and therefore should be deferred until disease activity is controlled with systemic therapy. Only local conservative débridement should be performed, if absolutely essential, for infected wounds because aggressive débridement leads to unnecessary soft-tissue damage and loss.
Skin-grafting has a high failure rate historically, and early reports suggested that skin grafting of ulcers was completely futile, with rejection occurring seven to thirteen days after the procedure5,6. More recently, the plastic surgery and dermatological surgery literature has described successful split-thickness skin-grafting of ulcers in patients with pyoderma gangrenosum7,8. The improved outcome of skin grafting follows recognition of the need for sustained preoperative stabilization of the ulcers with medical therapy, which maximizes the chances of a successful graft. Once the underlying disease process is controlled and the wound has been treated locally, reconstruction can be accomplished9.
We describe the case of a young patient who had acute bullous hemorrhagic pyoderma gangrenosum. The case described here illustrates the importance of a swift diagnosis to avoid a potentially catastrophic outcome. The patient and her family were informed that data concerning this case would be submitted for publication.
A twenty-one-year-old woman presented to her local emergency department with a painful, necrotic blister on the left leg, which had not responded to oral antibiotics. The patient thought that the blister may have followed an insect bite. She was referred to the orthopaedic surgeons, and the area was surgically débrided and successfully treated with skin-grafting. An abscess was diagnosed, although tissue specimens were not sent for microbiological analysis and blood cultures were negative. Three months later, she presented with a similar lesion on the right leg and was admitted to the same service with a presumed diagnosis of abscess. Examination revealed a large dusky blue blister in the middle part of the right calf, which rapidly evolved into an ulcer with a necrotic base and a bullous hemorrhagic border. Intravenous antibiotics were administered, and multiple surgical débridements were performed over a two-week period. Despite these interventions, the lesion rapidly advanced and the patient became systemically ill. The C-reactive protein level was 350 mg/L; the white blood-cell count was 35 × 109/L; blood film analysis, lymphocyte subsets, and serum protein electrophoresis revealed normal findings; and vasculitic and autoimmune screens and hepatitis serological testing were negative. Culture of blood and microbiological analysis of tissue specimens for bacteria, fungi, and mycobacteria were negative. The lesion continued to advance (Fig. 1-A) and the patient's general condition deteriorated; the admitting team suspected necrotizing fasciitis, and a below-the-knee amputation was contemplated. The extent of the systemic deterioration was such that the patient required transfer to the intensive care unit at another institution, where she was immediately referred to the dermatologists (E.L.N., M.M.B., and S.I.C.). Pyoderma gangrenosum was immediately suspected in view of the history of rapidly progressive noninfective ulceration that worsened after surgical débridement, combined with the clinical appearance of the lesion with its raised violaceous border and boggy necrotic base. Histological findings supported the diagnosis, with a diffuse neutrophilia and dermal neutrophilic microabscesses compatible with an untreated, expanding lesion. The patient was managed with pulsed intravenous methylprednisolone (1 g daily for three days), and substantial clinical improvement was evident after twenty-four hours. Treatment with oral cyclosporine (300 mg daily) and prednisolone (40 mg daily) was subsequently initiated. Clinical improvement was sustained for a period, and progression of the lesion was arrested. However, two weeks after the initiation of treatment, the ulcer edges showed development of purple-red borders and became increasingly painful, features highly suggestive of reactivation.
The patient received additional intravenous methylprednisolone (1 g daily for three days) as well as additional immunosuppression with the steroid-sparing agent mycophenolate mofetil (started at a dose of 1 g daily). Ultimately, a total of three cycles of methylprednisolone were required, with administration of prednisolone (40 mg daily), cyclosporine (300 mg daily), and mycophenolate mofetil (3 g daily) in order to sustain disease remission.
Extensive investigations, including computed tomography of the abdomen and pelvis and sigmoidoscopy, did not identify any underlying disease. After ten weeks of aggressive immunosuppressive therapy, the pyoderma gangrenosum was sufficiently stabilized to allow skin-grafting (Fig. 1-B). A good result was achieved, with 100% graft take and complete healing at the donor site. The patient was gradually weaned from all immunosuppressive therapy over twelve months, and there was no recurrence of disease during the nine-month period after the discontinuation of all treatment. The patient later sustained a metatarsal fracture that required orthopaedic surgical treatment with internal fixation, resulting in reactivation of the pyoderma gangrenosum at the surgical site on the right foot. Mycophenolate mofetil (2 g daily) was reinstated, and the ulceration healed after six weeks of therapy.
Typically, pyoderma gangrenosum begins as a small papule on the lower extremity, which the patient commonly attributes to an insect bite. The papule develops into a pustule that breaks down to form an ulcer with a purulent, necrotic base and an undermined, violaceous edge. Ulcerative lesions should be investigated for an underlying infectious cause, and biopsy and culture are mandatory. Bacterial and mycobacterial infections, deep mycoses, and other unusual infections as well as vasculitis and vascular insufficiency must be considered. However, pyoderma gangrenosum must be considered in the differential diagnosis of a rapidly expanding lesion that does not demonstrate growth of organisms on culture or respond to antibiotics, especially if there has been a similar problem at another site, as was the case in our patient.
This case highlights several important points regarding management. Pyoderma gangrenosum is often difficult to diagnose and challenging to treat. It may reveal itself in various clinical contexts, appearing unexpectedly after an elective surgical procedure as a wound that fails to heal or, as in the case of our patient, manifesting as idiopathic ulceration that worsens in spite of repeated surgical débridement.
When making the diagnosis, it is important to consider a history of similar lesions and to maintain a high index of clinical suspicion even in the absence of underlying systemic disease. A multidisciplinary approach with an early dermatological review is important, especially in difficult cases. The exaggerated and uncontrolled inflammatory response to nonspecific stimuli, resulting in pathergy, means that conservative local débridement should be performed only if deemed absolutely necessary. Inappropriate surgical interventions such as repeated extensive débridement will incite the disease process and cause additional tissue destruction.
There is now evidence to support a role for skin-grafting in the treatment of pyoderma gangrenosum, but this treatment modality is only useful when the disease is quiescent following prolonged immunosuppression. Randomized, blinded, prospective trials investigating the treatment of pyoderma gangrenosum are not available. The treatments that are associated with the best clinical evidence are systemic steroids and cyclosporine, with combinations of corticosteroids and agents such as azathioprine, cyclophosphamide, and mycophenolate mofetil being used for patients with recalcitrant disease10. The most effective dose and duration of immunosuppressive therapy remains to be determined8, but in practical terms the duration of therapy depends on the clinical response. The initial response is sometimes rapid, with epithelialization becoming evident within a few weeks after the start of treatment; however, remission must be sustained for an additional period of weeks before treatment is gradually tapered. In relatively straightforward cases, treatment may be phased out over several months. In difficult cases in which more than one immunosuppressant is required in order to obtain remission, or when skin-grafting is performed, treatment will invariably be more prolonged and immunosuppressant reduction must be cautious. Gilmour and Stewart reported a similar case of a severe, recalcitrant lesion that responded to cyclosporine and mycophenolate mofetil11. In that case, complete healing was observed after seven months and immunosuppression was gradually reduced at ten months without reactivation. Reactivation may occur in the context of tapering immunosuppressant therapy or as a result of a pathergic response to trauma; in either case, the clinical appearance of the lesion will indicate renewed disease activity (an erythematous or violaceous advancing border) along with increasing pain and elevated inflammatory markers. Our patient had further disease reactivation following surgery of the foot, despite the fact that the original lesions had remained quiescent after the discontinuation of all immunosuppressive treatment. This finding suggests that prophylactic steroids may be indicated for patients with a history of pyoderma gangrenosum if elective surgery is planned. Another point of interest in the case of our patient was the success of the split-thickness skin-grafting procedure that was performed at the time of the presentation, despite the absence of systemic immunosuppression. It is certainly rare for such sites to respond favorably to surgery without adjuvant therapy, and we can only speculate as to the cause of this unusual phenomenon. However, there have been other reported cases in the literature in which untreated lesions, predating the diagnosis of pyoderma gangrenosum, have healed spontaneously12,13. A diagnostic delay resulting in the late introduction of immunomodulatory therapy reduces the chances that surgical closure or grafting of wounds will be successful and increases the risk of extensive scarring and disfigurement. In the case of our patient, the implications of diagnostic delay were potentially disastrous, with limb amputation being under consideration before the correct diagnosis was made. Although there have been similar reports in the surgical literature, none have described a near-miss of such catastrophic proportions14. It is clearly imperative to identify and treat these lesions correctly in order to avoid massive tissue destruction and loss.
Acute pyoderma gangrenosum does not require surgical therapy; lesions first need to be stabilized with systemic immunosuppressive treatment and local wound care, which may include negative-pressure dressings and topical steroids to reduce slough and excessive granulation tissue, respectively. Minimal, careful débridement occasionally may be indicated when large amounts of necrotic or infected tissue are impeding the healing process. Surgical intervention should otherwise be restricted to patients with ulcers demonstrating healthy granulation tissue and minimal exudate who are receiving appropriate immunosuppression and do not have clinical signs of active disease. In these selected cases, the outcome of skin-grafting can be successful and is cosmetically superior to allowing the wound to heal by secondary intention. 
Hay CR, Messenger AG, Cotton DW, Bleehen SS, Winfield DA. Atypical bullous pyoderma gangrenosum associated with myeloid malignancies. J Clin Pathol.1987;40:387-92.40387
1987
[PubMed][CrossRef]
Hickman JG, Lazarus GS. Pyoderma gangrenosum: a reappraisal of associated systemic diseases. Br J Dermatol.1980;102:235-7.102235
1980
[CrossRef]
Weenig RH, Davis MD, Dahl PR, Su WP. Skin ulcers misdiagnosed as pyoderma gangrenosum. N Engl J Med.2002;347:1412-8.3471412
2002
[CrossRef]
Prystowsky JH, Kahn SN, Lazarus GS. Present status of pyoderma gangrenosum. Review of 21 cases. Arch Dermatol.1989;125:57-64.12557
1989
[CrossRef]
Long PI, Uesu CT. Pyoderma gangraenosum. JAMA.1964;187:336-9.187336
1964
Wustrack KO, Zarem HA. Pyoderma gangrenosum: recognition and management. Plast Reconstr Surg.1978;62:423-8.62423
1978
[CrossRef]
Zakhireh M, Rockwell WB, Fryer RH. Stabilization of pyoderma gangrenosum ulcer with oral cyclosporine prior to skin grafting. Plast Reconstr Surg.2004;113:1417-20.1131417
2004
[CrossRef]
Cliff S, Holden CA, Thomas PR, Marsden RA, Harland CC. Split skin grafts in the treatment of pyoderma gangrenosum. A report of four cases. Dermatol Surg.1999;25:299-302.25299
1999
[CrossRef]
Hansen SL, Mathes SJ. Problem wounds and principles of closure. In: Mathes SJ. Plastic surgery. 2nd ed. Philadelphia: Saunders; 2005. p 940.
2005
Reichrath J, Bens G, Bonowitz A, Tilgen W. Treatment recommendations for pyoderma gangrenosum: an evidence-based review of the literature based on more than 350 patients. J Am Acad Dermatol.2005;53:273-83.53273
2005
[CrossRef]
Gilmour E, Stewart DG. Severe recalcitrant pyoderma gangrenosum responding to a combination of mycophenolate mofetil with cyclosporine and complicated by a mononeuritis. Br J Dermatol.2001;144:397-400.144397
2001
[CrossRef]
Wong CK, Pun KK, Lum CC, Lee SW, Ng MM, Wang CC. Pyoderma gangrenosum associated with erythroid hypoplasia. Postgrad Med J.1990;66:312-3.66312
1990
[CrossRef]
Brown RE, Lay L, Graham D. Bilateral pyoderma gangrenosum of the hand: treatment with dapsone. J Hand Surg [Br].1993;18:119-21.18119
1993
Bennett CR, Brage ME, Mass DP. Pyoderma gangrenosum mimicking postoperative infection in the extremities. A report of two cases. J Bone Joint Surg Am.1999;81:1013-8.811013
1999
[CrossRef]