A 33-year-old man presented with a wound on the right lower leg that had gradually gotten worse over the past month. The patient had treated the wound at home with peroxide and triple antibiotic ointment. He could not recall being injured. His only symptoms were pain at the wound site and swelling of his right foot.
The patient’s medical history included protein S deficiency. Although he had no history of pulmonary embolism, he had experienced 7 episodes of deep venous thrombosis (DVT) over the past 10 years. There were no signs or symptoms of diabetes, coronary artery disease, or hypertension and there was no family history of venous thromboembolic disease. The patient was a current smoker with a 15-pack-year history and also suffered from depression. He had been taking an anticoagulant for 10 years for protein S deficiency and also was taking an antidepressant.
The wound site comprised a cluster of lesions filled with yellow fibrinous exudate, slough, and proteinaceous debris, surrounded by eschar. There was full-thickness skin loss and the entire area was erythematous and swollen (Figure). Circumference measurements taken of the patient’s lower legs were: right calf, 41.5 cm; right ankle, 23 cm; left calf, 37 cm; and, left ankle, 21 cm. Dorsal pedis and posterior tibial pulses were 2+ on both feet. The patient had no tenderness in his calf and he was not in distress. Findings from the rest of the physical examination was normal.
Laboratory test results for complete blood cell count, HbA1C, and protein status were within normal limits. The patient’s INR was 2.6 (normal range, 2.0 to 3.0). Laboratory values were elevated for inflammatory markers (ESR and CRP). A lower extremity arterial Doppler study found no evidence of DVT. Results of a deep wound culture were positive for methicillin-resistant Staphylococcus aureus. Results of a tissue biopsy suggested warfarin-induced skin necrosis.
Warfarin was immediately discontinued and low molecular weight heparin (LMWH) was substituted. A 14-day course of sulfamethoxazole and trimethoprim DS was initiated and the patient was instructed in the use of silver-based antimicrobial dressings. He was referred to the wound care clinic for weekly excisional debridements.
Initially, the wound responded well to the discontinuation of warfarin. Level of bioburden decreased significantly and healthy granulation tissue began to form at the lesion site. Unfortunately, secondary infections were persistent and healing was set back. For several months, the patient continued a regimen of debridement, topical antimicrobial dressings, and oral antibiotics when necessary to treat infection. The wound eventually healed, but the patient subsequently had a pulmonary embolism while receiving LMWH. He subsequently received an ICV filter.
Warfarin-induced skin necrosis affects 0.01% to 0.1% of patients taking the anticoagulant and more than 300 cases have been reported worldwide.1 The clinical presentation of warfarin-induced skin necrosis often starts with paresthesia, sensations of pressure, and edema and is followed by petechiae that progress to well-demarcated bullae. The lesions are acutely painful and-if the loading dose of warfarin is high-may appear within the first 24 hours of therapy. The majority of cases appear between day 3 and 6 of therapy. Bullae become hemorrhagic and quickly progress to necrotic eschar, typically within 7 to 10 days after initiation of therapy.2 The eschar ultimately sloughs, revealing a full-thickness wound extending into subcutaneous fat. Some patients require extensive surgical debridement.1
Warfarin-induced skin necrosis is not limited to the first course of therapy. Case reports indicate that lesions may appear after subsequent treatments and even months to years after initiation of therapy.3 Diagnosis relies on presentation, history of warfarin use, and histopathologic findings. The differential diagnosis of acute skin necrosis includes necrotizing fasciitis, disseminated intravascular coagulation, heparin induced thrombosis, cellulitis, Fournier’s gangrene, pyoderma gangrenosum, and calciphylaxis.
Warfarin-induced skin necrosis occurs most frequently in middle-aged, perimenopausal, obese women being treated for DVT or pulmonary embolism.1 Classically-and unlike the patient in this case-lesions appear in areas with significant subcutaneous fat, most notably the breast, buttock, abdomen, or thigh. Although the condition can occur in both men and women, there are no reports in the literature describing typical presentations in men. Thrombophilia is a predisposing factor, as are conditions that require anticoagulation, such as deficiencies in factor VII, protein C, and protein S.
Unfortunately, little is known about histologic changes that precede tissue necrosis. Biopsies are usually taken later in the necrotic process. Nalbandian and colleagues4 suggest that warfarin may injure capillaries at the dermovascular loop. The drug-damaged vessels dilate and rupture and petechiae develop quickly. Venules and veins distal to the injury thrombose, blood supply to the surrounding tissue is cut off, and necrosis begins. The exact pathogenesis is still not well understood.
Patient complaint of localized skin discomfort may be the first sign of the condition and should prompt immediate discontinuation of warfarin and replacement with an alternative anticoagulant to prevent DVT and pulmonary embolism. A vascular filter may be needed if anticoagulation fails. Even in early stages of skin breakdown, excisional debridement is often required. Skin grafting may sometimes be necessary to repair large, non-healing wounds. Amputation of an affected extremity is rare.
• Warfarin-induced skin necrosis is a rare, but well recognized, complication of warfarin therapy.
• The differential diagnosis should rule out other causes of acute skin necrosis (eg, necrotizing fasciitis, disseminated intravascular coagulation, heparin induced thrombosis, cellulitis, Fournier’s gangrene, pyoderma gangrenosum, and calciphylaxis).
• Early diagnosis and immediate cessation of warfarin therapy can halt progression to, or limit the extent of cutaneous necrosis.
• Occurrence of warfarin-induced skin necrosis should not outweigh the benefit of treatment with warfarin; the risk of not treating a coagulopathy is greater than the very small possibility of skin necrosis.
References:1. Chan YC, Valenti D, Mansfield AO, Stansby G. Warfarin induced skin necrosis. Br J Surgery. 2000; 87:266-272.
2. Nazarian RM, Van Cott EM, Zembowicz A, Duncan LM. Warfarin-induced skin necrosis. J Am Acad Dermatol. 2009;61:325-332.
3. Ward CT, Chavalitanonda N. Atypical warfarin-induced skin necrosis. Pharmacother. 2006; 26:1175-1179.
4. Nalbandian RM, Mader IJ, Barrett JL, et al. Petechiae, ecchymoses, and necrosis of skin induced by coumarin congeners: rare, occasionally lethal complications of anticoagulant therapy. JAMA. 1965;192:107-112.