Pyoderma Gangrenosum: A Misdiagnosis

February 2013 | Volume 12 | Issue 2 | Case Reports | 228 | Copyright © February 2013

Jessica El-Kehdy MD,a Eckart Haneke MD,b and Paula G. Karam MDc

aAmerican University of Beirut Medical Center, Beirut, Lebanon bDermatol Practice Dermaticum, Freiburg, Germany cBalamand University, Beirut, Lebanon

Pyoderma gangrenosum is an inflammatory disease of unknown etiology, commonly mistaken for an infection. Here, we report the case of a 55-year-old woman with an infectious process that was initially misdiagnosed as pyoderma gangrenosum and treated accordingly. We also discuss the criteria and the methods available to diagnose infections and pyoderma gangrenosum correctly.

J Drugs Dermatol. 2013;12(2):228-230.


Postoperative pyoderma gangrenosum (PG), also known as postoperative progressive gangrene of Cullen,1 is a rare complication of surgical interventions that is usually mistaken for an infection. Here, we report a case of a probable infection that was initially misdiagnosed as PG because of poor wound healing in response to early antibiotic treatment.


A 55-year-old woman with a known history of dyslipidemia who was on fluvastatin therapy 80 mg once daily underwent a scar revision of the periareolar area. The patient had undergone breast reconstruction surgery 4 years prior, with a technique involving a cut only in the periareolar area with undermining of tissues and deep sutures to hold the mammary glands. Five days postoperatively, she developed erosions and ulcerations that enlarged. The diagnosis made by the surgeon was an inflammatory reaction to the sutures, and only local treatment was applied. However, this treatment remained ineffective.
Five months later, the ulcer had healed accidentally, leaving a scar, after the patient took systemic antibiotics for impetigo of the skin. One year later, the patient consulted the plastic surgeon again, and a scar revision for both breasts was performed. The patient healed normally. Another 2 years later, another scar revision was done to remove the rest of the remaining scar tissues. Four days postoperatively, the patient noticed a black discoloration on the left breast at the area of the suture. On physical examination, necrosis was noted (Figure 1a) that soon turned into an ulcer that began to grow in size (Figures 1b and c), although the patient was already on oral ciprofloxacin therapy at 500 mg twice daily. The surgeon debrided the area, but the ulcer continued to increase in size, with a rolled-in border. The patient's symptoms were minimal, and she described her pain as lasting only few minutes and directly preceding an increase in the size of the lesion. Several bacterial cultures were negative.
After consulting several dermatologists, the diagnosis of PG was made based on the clinical picture. A biopsy was not performed for fear of exacerbating the condition due to the pathergy effect. Oral prednisone was started at 1.5 mg/kg/day for 1 week and then tapered slowly over another week. While on steroids, the ulcer continued to increase in size, and a new ulcer appeared on the right breast. Diagnostic tests were done to rule out any of the conditions usually associated with PG, including leukemia and inflammatory bowel diseases. All laboratory test results were negative except for an increase in serum β-globulin and γ-globulin (β-globulin, 15.3%, normal range, 8.9%-13.6%; γ-globulin, 19.3%, normal range, 8.4%-18.3%), a decrease in serum albumin (49.4%, normal range, 53.0%-74.0%), a slight decrease in albumin/globulin ratio (0.98, normal range, 1.0-2.5), and a slight increase in immunoglobulin G antigliadin antibodies (23.3 U, negative value, <20 U).
However, a work-up was done to check for the presence of celiac disease and multiple myeloma, but all results were negative and these diseases were excluded. Bacterial cultures were repeated and showed Klebsiella species and Escherichia coli with poor growth and rare Staphylococcus epidermidis. At the same time, a skin biopsy with hematoxylin and eosin staining under medium power magnification was performed that showed an ulcer with no evidence of PG. Empiric oral fluconazole was started at 150 mg/day for 8 days in addition to intravenous meropenem at 500 mg every 8 hours for 8 days. Four days after the initiation of therapy, the ulcer on the left breast began to regress in size and heal (Figure 1d). However, on the right breast, a thin thread of tissue was seen protruding, and when cut by the surgeon, the area healed, leaving scars.


First described by Brunsting et al in 1930,2 PG is an inflammatory disease of unknown etiology, belonging to a spectrum of disorders known as neutrophilic dermatoses ,3 which have in common a predisposition to pathergy and the presence of noninfectious neutrophilic infiltrates, mostly in the skin.4 PG begins as a painful, erythematous nodule that turns into a tender ulcer with a swollen, asymmetrical, undermined border and a necrotic hemorrhagic base.5 Roughly 50% of cases are associated with conditions such as inflammatory bowel diseases, leukemia, and rheumatoid arthritis, while the remaining 50% are idiopathic.