Pyoderma gangrenosum is characterized by the development of necrosis and ulcerations with typical purple undermined edges. Ulcers are either single or multiple and occur primarily on the lower extremities but can appear anywhere. No infectious cause has been identified. Early experience with debridement and grafting of pyoderma gangrenosum ulcers yielded disappointing, sometimes catastrophic, results.35 This was mostly the result of the presence of pathergy, a key feature in the disease process, in which any traumatized skin (debridement sites or skin graft donor sites) develop additional necrosis and ulceration. 30 For some time plastic surgeons have shied away from treating these patients surgically. This attitude has persisted, until recent new insights in the pathophysiology of the disease process and new therapeutic regimens prompted a change in the management of select patients in our institution. 9–12 We resort to debridement and skin grafting only after controlling the underlying disease process with immunosuppressive drugs to decrease morbidity, the period of wound care, and the length of the hospital stay. Postoperatively, very slow tapering of corticosteroids is deemed essential to prevent recurrence. We review our experience with 4 patients who presented with large ulcers.
Materials and Methods
Charts of patients diagnosed with pyoderma gangrenosum from January 1972 to December 1999 were reviewed. Age, sex, demographic data, associated underlying disease, and ulcer characteristics such as size, number, location, treatment type, dose, and duration are presented in the Table. The healing period from time of onset or presentation until complete healing of all wounds is noted along with the type of therapeutic treatment.
Twenty-two charts were reviewed for patients with the clinical diagnosis of pyoderma gangrenosum. Ten charts were retained because of sufficient documentation and a well-documented diagnosis. Patient characteristics are presented in the Table.
Only 4 patients with large ulcers (>5 × 5 cm) treated with debridement and split-thickness skin grafting were brought to the attention of the senior author. No other patients underwent a surgical procedure; they were treated completely without surgical consultation. All surgically treated patients were discharged within 1 week of skin graft application and complete healing of their ulcers. The following is a summary of the 4 patients with large ulcers in whom surgery was offered after medical therapy controlled the progression of the disease.
The first patient is a 13-year-old girl with inflammatory bowel disease (ulcerative colitis) who presented with a 2-week history of an 8 × 6-cm necrotic, mildly painful ulcer on her right lateral malleolar area and another 3 × 2-cm ulcer on the dorsum of her left foot. The ulceration was not related to reactivation of her bowel disease. She was started for 3 weeks on 15 mg Solu-Medrol twice daily followed by skin grafting of her lesions. Corticosteroid therapy was continued at a dosage of 30 mg Prednisone per day for 2 weeks, and was then tapered slowly. The patient was discharged 1 week postoperatively. A 3-mm purplish pustule at the lower edge of her right ankle appeared 2 months later (Fig 1A) and was treated successfully with boluses of steroids (Fig 1B). The patient was disease free 8 months postoperatively.
The second patient is a 15-year-old boy with no concurrent illnesses who developed a 10 × 6-cm ulcer with purplish edges on his right lateral malleolus and another 3 × 2-cm ulcer on his left medial malleolus. Two cultures were negative. Biopsy showed neutrophilic infiltrates, and his clinical picture was compatible with pyoderma gangrenosum. He was treated for 8 weeks with 50 mg Prednisone daily, tapered by 5 mg per week. When he presented to our office, his left ankle ulcer had healed and he had a residual 8 × 6-cm ulcer on his right ankle. A skin graft was applied on an outpatient basis, and the patient did well afterward. No recurrence was noted at 6 months nor was there any other manifestation of underlying disease.
The third patient is a 32-year-old man who developed acute, diffuse vasculitis with renal, pulmonary, and intestinal involvement. Progressive cutaneous ulcerations appeared, with negative cultures that were refractory to broad-spectrum antibiotics. Skin biopsy demonstrated neutrophilic infiltrates compatible with pyoderma gangrenosum. The patient’s clinical picture improved spontaneously over 3 weeks, and he underwent skin grafting for a 5 × 8-cm ulcer on his right foot, a 3 × 3-cm ulcer on his right leg, and a 6 × 4-cm ulcer on his left arm with complete healing. The patient was discharged 1 week later.
The fourth patient is a 56-year-old man with a 3-week history of painless pustules (Fig 2A) progressing to coalescent ulcers over the right and left leg areas (Fig 2B). He had an 8 × 6-cm ulcer on his left leg and two ulcers on his right leg that measured 8 × 5 cm and 7 × 5 cm respectively. All cultures were negative. The patient was treated with multiple local and systemic antibiotics to no avail. Two biopsies revealed neutrophilic infiltrates and no organisms—compatible with pyoderma gangrenosum. The patient was started on Prednisone 60 mg per day, and on day 5 underwent debridement and application of a split-thickness skin graft. The steroids were tapered at 6 months. The patient had an uneventful course with stable coverage at 14 months. Figure 2C shows stable coverage at 2 months.
The other 6 patients were treated medically. Steroids were the only immunosuppressors used, in addition to the local wound care (saline dressing and topical antibiotics). Patient 5 had a pustular form of the disease associated with systemic lupus erythematosus, and her lesions healed after 2 weeks of corticosteroid therapy. Ulcers treated without steroids by local wound care only took 5 to 8 months to heal. Ulcers treated with corticosteroid therapy took 3 to 6 months to heal. One patient (Patient 8) required high doses of steroids for 1.5 years, which resulted in the side effects of such a treatment, such as osteoporosis and cushingoid facies.
The diagnosis of pyoderma gangrenosum is a clinical one. 7 It is characterized by a necrotic ulcer that is typically undermined with purple edges. It is single in 52% of patients. A total of 11% of patients have more than five ulcers, and pain is present in 58% of patients 1–4. The lower legs and feet are the site of predilection. 8–11 Cultures rule out a primary infectious process, and a response to immunosuppressors is usually expected. Pathologically, an occlusive process in the dermal and subdermal vessels is seen, with a polymorphonuclear infiltrate surrounding small-caliber vessels. Sometimes, a lymphocytic infiltrate is seen. 12,13 An abnormality of leukocyte function has been demonstrated, 14–19 showing either a deficient or an excessive but ineffective margination. Some authors 16,17 have demonstrated an abnormality in the interaction between leukocyte membrane urokinase receptors (uPAR) and β-2 integrins CR3 and CR4. These integrins were overexpressed and clustered 17. Phenotypically, this interaction is manifested by a faster and more chaotic oscillation in its pattern when compared with normal migrating neutrophils. This pattern is restored to normal with subsaturating doses of phosphatase inhibitors or with a pulsed electric field. 16–17N-acetyl-d-glucosamine disrupted integrin clustering 17. A nicotinamide adenine dinucleotide phosphate (hydrogen) oscillation abnormality has been linked previously to abnormal cell trafficking. 25,26
Pyoderma gangrenosum is associated with an underlying disease in 50% to 67% of patients 10 —most commonly inflammatory bowel disease, and hematological and lymphatic malignancies. Immunoglobulin (Ig) A or IgM monoclonal gammopathy, hepatitis C or other viral infections, rheumatoid arthritis, Wegener’s granulomatosis, systemic lupus erythematosus, and other autoimmune diseases are other cited causes. 17,20 It has also been associated recently with human immunodeficiency virus. 21
The mainstay of therapy is high doses of corticosteroids initially, followed by slow tapering to prevent recurrence. When steroids fail initially, cyclosporin A has been used successfully. 5,22,23 Tacrolimus (FK506) has also been used in case of steroid failure, 15,23,24 but we prefer to keep it as a second line of treatment after cyclosporin A. Methotrexate has been useful in chronic cases in patients who cannot be weaned off steroids. 6 Other agents that have been used include Dapsone, minocycline, hyperbaric oxygen, cyclophosphamide, mycophenolate mofetil, granulocyte-colony stimulating factor locally, sodium chromoglycate, and lymecycline with topical benzoyl peroxide. 25–28
Isolated attempts at surgical therapy have been disappointing. Additional necrosis has appeared at sites of debridement, and new ulcers have appeared at skin graft donor sites or surgical wounds locally or distantly. This is called pathergy. Plastic surgeons have traditionally shied away from treating these patients, and thus internists, pediatricians, and dermatologists have since handled them.
Because many resultant ulcers are large, waiting for these wounds to heal by secondary intention with medical therapy may subject these patients to unnecessary emotional and physical scarring, which would also increase their morbidity and hospital stay. Therefore, surgery may be a helpful adjunctive tool once medical therapy has controlled the progression of the disease. Our understanding of the disease process and the management of clinical ulcers have expanded.
We explored the possibility of skin grafting in these patients under the following provisions:
- The initial inflammatory and necrotic phases have been controlled medically with sufficiently high doses of steroids continued through the time of surgery.
- Slow tapering of the immunosuppressors is performed for a 6-month period to prevent recurrence.
We report very limited experience with adjunctive surgical management. Patient 1 had rapid tapering of her steroids after skin grafting and developed early recurrence of pyoderma gangrenosum at the edge of the skin graft. It is clear that the underlying autoimmune process was still continuing despite clinical healing of the ulcers. Untreated patients in our series, or patients who had only local wound care healed in 5 to 8 months. It seems that the underlying autoimmune process was self-limited to this period of time in a good number of our patients. This observation, in conjunction with other reports in the literature advocating slow tapering of steroids, 30 suggests that a period of immunosuppressive coverage of a few months (approximately 6 months) postoperatively is reasonable. This will avoid local recurrence at the site of surgery and the de novo appearance of distant new ulcers, which should be a pivotal part of any surgical plan.
We applied 2 to 3 weeks of immunosuppressive therapy to all our surgical patients. Only Patient 4 had a shorter course of immunosuppressive therapy (5 days) with no complications reported. However, we still favor a longer course of immunosuppression. We could not find reports in the literature supporting this particular issue; however, we have observed that the aggressive necrotic phase lasted approximately 2 to 3 weeks with steroid treatment, and there was a build-up of healthy granulation tissue thereafter. This particular issue deserves further investigation.
We have found very few other reports in the literature that support skin grafting for these lesions. All are based on anecdotal reports, and emphasize the importance of controlling the underlying disease. 28,29 Cliff and colleagues 30 reported four patients with pyoderma gangrenosum who underwent skin grafting under immunosuppression following a rationale similar to ours, and their experience supports our conclusions. Perioperative pyoderma gangrenosum is also an issue for plastic surgeons and general surgeons alike in handling these patients. Pathergy manifests when minor or major procedures are indicated in these patients, and even percutaneous access sites can develop into a clinical ulcer.31 Strategies described in the literature to minimize perioperative pathergy include operating during a quiescent phase and using subcuticular sutures rather than sutures going through the skin.32 We avoided intervening during active phases when the disease was not controlled medically.
We present our experience with select patients in closing large wounds with the goal of decreasing morbidity and hospital stay for these unfortunate patients. Pyoderma gangrenosum can be managed by a team of medical and surgical physicians. The patients are induced initially into a phase of remission, as evidenced by a cessation of further necrosis. They then undergo skin grafting. The length of preoperative and postoperative corticosteroid therapy necessary for complete remission are two issues that deserve additional investigation. This is probably better addressed through a multicenter study that evaluates managed patients with this rare entity.
1. Wustrack KO, Zarem HA. Pyoderma gangrenosum: recognition and treatment. Plast Reconstr Surg 1978; 62(3): 423–428
2. Cliff S, Holden CA, Thomas PR, et al. Split skin grafts in the treatment of pyoderma gangrenosum: a report of four cases. Dermatol Surg 1999; 25(4): 299–302
3. Dowden RV. Skin grafting of pyoderma gangrenosum lesions. Plast Reconstr Surg 1987; 80(4): 648[Letter]
4. Havlik RJ, Giles PD, Havlik NL. Pyoderma gangrenosum of the breast: sequential grafting. Plast Reconstr Surg 1998; 101(7): 1909–1914
5. Goldberg NS, Ottuso P, Petro J. Cyclosporine for pyoderma gangrenosum. Plast Reconstr Surg 1993; 91(1): 91–93
6. Teitel AD. Treatment of pyoderma gangrenosum with methotrexate. Cutis 1996; 57(5): 326–328
7. Goulden B, Bond L, Highet AS. Pyoderma gangrenosum associated with paroxysmal nocturnal hemoglobinuria. Clin Exp Dermatol 1994; 19(3): 271–273
8. Callen JP. Pyoderma gangrenosum and related disorders. Med Clin North Am 1989; 73(5): 1247–1261
9. Prystowsky JH, Kahn SN, Lazarus GS. Present status of pyoderma gangrenosum. Review of 21 cases. Arch Dermatol 1989; 125(1): 57–64
10. Von den Driesch P. Pyoderma gangrenosum: a report of 44 cases with follow-up. Br J Dermatol 1997; 137(6): 1000–1005
11. Powell FC, Su WP, Perry HO. Pyoderma gangrenosum: classification and management. J Am Acad Dermatol 1996; 34(3): 395–409
12. Lear JT, Atherton MT, Byrne JP. Neutrophilic dermatoses: pyoderma gangrenosum and Sweet’s syndrome. Postgrad Med J 1997; 73(856): 65–68
13. Davies MG, Hastings A. Sweet’s syndrome progressing to pyoderma gangrenosum—a spectrum of neutrophilic skin disorders in association with cryptogenic cirrhosis. Clin Exp Dermatol 1991; 16(4): 279–282
14. Dwarakanath AD, Yu LG, Brookes C, et al. ‘Sticky’ neutrophils, pathergic arthritis, and response to heparin in pyoderma gangrenosum complicating ulcerative colitis. Gut 1995; 37(4): 585–588
15. Abu-Elmagd K, Van Thiel DH, Jegasothy BV, et al. Resolution of severe pyoderma gangrenosum in a patient with streaking leukocyte factor disease after treatment with tacrolimus (FK-506). Ann Intern Med 1993; 119(7): 595–598
16. Adachi Y, Kindzelskii AL, Cookingham G, et al. Aberrant neutrophil trafficking and metabolic oscillations in severe pyoderma gangrenosum. J Invest Dermatol 1998; 111(2): 259–268
17. Shaya S, Kindzelskii AL, Minor J, et al. Aberrant integrin (CR4; alpha(x)beta2; CD11c/CD1) oscillations on neutrophils in a mild form of pyoderma gangrenosum. J Invest Dermatol 1998; 111(1): 154–158
18. Bedlow AJ, Davies EJ, Moss AL, et al. Pyoderma gangrenosum in a child with congenital partial deficiency of leucocyte adherence glycoproteins. Br J Dermatol 1998; 139(6): 1064–1067
19. Ross HJ, Moy LA, Kaplan R, Figlin RA. Bullous pyoderma gangrenosum after granulocyte colony-stimulating factor treatment. Cancer 1991; 68(2): 441–443
20. Handfield-Jones SE, Parker SC, Fenton DA, et al. Wegener’s granulomatosis presenting as pyoderma gangrenosum. Clin Exp Dermatol 1992; 17(3): 197–200
21. Paller AS, Sahn EE, Garen PD, et al. Pyoderma gangrenosum in pediatric acquired immunodeficiency syndrome. J Pediatr 1990; 117(1): 63–66
22. Matis WL, Ellis CN, Griffiths CE, Lazarus GS. Treatment of pyoderma gangrenosum with cyclosporine. Arch Dermatol 1992; 128(8): 1060–1064
23. D’Inca R, Fagiuoli S, Sturniolo GC. Tacrolimus to treat pyoderma gangrenosum resistant to cyclosporin. Ann Intern Med 1998; 128(9): 783–784
24. Reich K, Vente C, Neumann C. Topical tacrolimus for pyoderma gangrenosum. Br J Dermatol 1998; 139(4): 755–757
25. Nousari HC, Lynch W, Anhalt GJ, Petri M. The effectiveness of mycophenolate mofetil in refractory pyoderma gangrenosum. Arch Dermatol 1998; 134(12): 1509–1511
26. Anderson LL, Samlaska CP, Cardone JS, Holtzmuller KC. Treatment of pyoderma gangrenosum with 4% cromolyn. Arch Dermatol 1994; 130(9): 1117–1120
27. Davis JC, Landeen JM, Levine RA. Pyoderma gangrenosum: skin grafting after preparation with hyperbaric oxygen. Plast Reconstr Surg 1987; 79(2): 200–207
28. Vereecken P, Wautrecht JC, De Dobbeleer G, Heenen M. A case of pyoderma gangrenosum stabilized with lymecycline, topical benzoyl peroxide and treated by autograft. Dermatology 1997; 195(1): 50–51
29. Gooding JM, Kinney TB, Oglevie SB, Wang-Rodriguez J. Pyoderma gangrenosum twice complicating percutaneous intervention in a single patient. AJR Am J Roentgenol 1999; 172(5): 1352–1354
30. Long CC, Jessop J, Young M, Holt PJ. Minimizing the risk of post-operative pyoderma gangrenosum. Br J Dermatol 1992; 127(1): 45–48