Necrotic Drug eruptions (Heparin-induced skin necrosis; warfarin-induced skin necrosis (Coumadin® necrosis)

Are You Confident of the Diagnosis?

Heparin and warfarin-induced skin necrosis need to be considered in this category as drug eruptions that cause dermal and epidermal necrosis. These two topics are covered in detail in this chapter. Toxic epidermal necrolysis and Stevens-Johnson syndrome cause epidermal necrosis. They will be considered in a separate chapter, so named.


What you should be alert for in the history

A history of heparin administration is essential in making this diagnosis. The patient may be receiving this intravenously, subcutaneously, or in small amounts for line flushes during dialysis or other infusion lines. Heparin necrosis has also been reported after injection of low-molecular-weight (LMW) heparins. Initially, there is pain and erythema at heparin injection sites. There may also be edema and bulla formation. Gradually, the affected areas become necrotic. This can usually take 5 to 9 days to occur.

Characteristic findings on physical examination

Necrotic plaques have a retiform appearance, which suggests their intravascular pathogenesis. Lesions may appear earlier in the course of treatment if heparin has been given previously. Necrotic plaques may also develop at sites distant to injection sites. Plaques vary from 2 to 10 cm, but occasionally, plaques up to 15 cm wide have been reported. A review of 21 cases associated with various preparations of LMW heparins showed that the average time to the development of necrosis was 7.6 days. Necrosis occurred both at the injection site and at sites distant to the injection site in 24% of patients.

Expected results of diagnostic studies

The diagnosis of heparin-induced skin necrosis is made based on the history of heparin administration in conjunction with the clinical picture. Histopathologic findings and serological tests confirm the diagnosis and differentiate it from other retiform necrotic disorders

Histopathologic findings show intravascular thrombi within superficial vessels in the dermis and resultant epidermal and dermal necrosis. A secondary leukocytoclastic vasculitis may also be seen.

The importance of making this diagnosis is that it may be a marker for an underlying immune complex disorder to heparin, known as type 2 heparin-induced thrombocytopenia (HIT). Therefore, this diagnosis should prompt an evaluation of the platelet count, which will be low in most but not all cases. If the platelet count is normal, serological evidence of HIT should be sought by screening for anti-heparin/platelet factor 4 (PF4) antibodies by immunoassay and then obtaining the more specific platelet activation assay.


Characteristic findings on physical examination

The initial presentation is that of edema, pain, or parasthesiae in involved sites. Purpuric macules are followed by the appearance of hemorrhagic bullae and frank necrosis (Figure 1). Pain may be severe, especially when frank necrosis is present. Areas with ample subcutaneous fat are those most commonly affected, including the buttocks, breasts, abdomen, and thighs. Other sites may be involved, such as the legs, ams, back, and penis. Eyelid skin involvement has been reported.

Figure 1.

Coumadin necrosis.

The median time to onset is 5 days after drug initiation (range 3 to 10 days). A smaller cohort of patients in the literature have presented with a late-onset necrosis occuring months to years of starting warfarin. It may also be seen in patients who have tolerated warfarin previously, in whom therapy is re-initiated. Cases occuring within days of discontinuation of warfarin have also been reported.

Expected results of diagnostic studies

Histopathologically, there is ischemic epidermal and dermal necrosis, which arises secondary to plugging of vessels by microthrombi. Capillaries, venules and deep veins may be affected.


The diagnosis of HISN and WISN may be straightforward in typical cases with a typical history. However, clinical presentations vary, and the following conditions need to be borne in mind

Calciphylaxis may be challenging to distinguish from WISN clinically. This dilemma may arise especially in patients with chronic renal failure. However, calciphylaxis has been reported in patients without renal insufficiency as well. In these patients, obesity, diabetes, hyperlipidemia and a history of smoking are usually present. Women are more frequently affected.

In calciphylaxis, retiform purpura progesses to retiform necrosis. Plaques are deep and tender and usually occur on the thighs. The pannus and lower legs are also commonly involved. Histopathologically, the classic findings are calcium deposition in lobular capillaries and small arterioles in the fat. Calcium may also be seen lying within the dermis and subcutaneous fat.

Disseminated intravascular coagulopathy. Here, confluent retiform purpura eventuate into necrotic plaques in the setting of an acutely ill patient. The plaques are more widespread and superficial than in WISN.

Other causes of retiform purpura and necrosis also need to be borne in mind, including other thrombotic disorders (antiphosholipid syndrome, antithrombin III deficiency) , embolic disorders (septic emboli, cholesterol emboli), systemic sepsis, and medium-vessel vasculitis.

Necrotizing fasciitis also presents with necrosis. A diagnostic clue in the early stages is systemic findings (high fever, an unwell patient with severe pain) out of proportion to skin findings (may be minimal in the early stages with a small area of necrosis, which will ultimately progress).

Who is at Risk for Developing this Disease?

Heparin-induced skin necrosis is a rare side effect of heparin therapy. The incidence of this complication varies between 1% and 3% for a 1-week course of intravenous heparin therapy. Elderly females receiving heparin are the group most commonly reported.

The incidence of LMW heparin-induced skin necrosis is reported to be lower than that to unfractionated heparin. A slight female predominance has also been reported in this group. While heparin necrosis has not been reported after heparin flushes of indwelling catheters, HIT has been reported after hemodialysis.

Warfarin-induced skin necrosis is also a rare phenomenon. It is estimated to occur in 1 in 1000 to 1 in 10,000 patients receiving warfarin anticoagulation. Middle-aged overweight females receiving warfarin for thrombophlebitis or pulmonary embolism are those most commonly affected. High loading doses of warfarin and inadequate heparinization prior to warfarin initiation have been reported as risk factors for WISN.

What is the Cause of the Disease?

Heparin-induced skin necrosis occurs most frequently in the context of type 2 heparin-induced thrombocytopenia (HIT). This syndrome is defined as a relative decrease of the platelet count to below 50% of its baseline, or an absolute decrease to less than 100×109/L. In this syndrome, heparin-induced anti-platelet IgG antibodies are formed when heparin binds to a component of the alpha- granules of platelets, PF4. The resultant antigen-antibody complex adheres to and activates platelets, which leads to the formation platelet thrombi and occlusion of vessels. Platelet counts may appear normal, and so the syndrome may not be readily apparent in cases where previous platelet counts have not been performed.

While it is suspected to be high, the true incidence of type 2 HIT in these patients is unknown. The newer literature refers to heparin-induced skin necrosis as a localized form of type 2 HIT. On the other hand, heparin- or LMW heparin-induced skin necrosis has been reported to occur in about 10% to 20% of patients who develop systemic HIT after these subcutaneous injections. Also, arterial thrombosis has been reported to occur at higher rates in those patients with heparin-induced skin necrosis who develop low platelets when compared with those who don’t.

In the LMW heparin skin necrosis study mentioned above, thrombocytopenia was seen in 9 of 19 patients. Only 10 of 21 patients had been screened for anti-heparin/PF4 platelet antibodies, and 9 of 10 patients were positive. A single patient with severe thrombocytopenia was tested more specifically with the platelet activation test, which was negative (anti-heparin/PF4 platelet test not performed). Therefore, it is still possible that another pathomechanism may be operable is some patients.

In WISN, warfarin may paradoxically induce a transient, relatively procoagulant state. Inherited or acquired protein C deficiency predisposes to WISN, although most people with protein C deficiency tolerate warfarin, so other factors must be operable. An association with other clotting syndromes, such as antithrombin III deficiency, the antiphospholipid syndrome and factor V Leiden mutations, as well as protein S deficiency, has been reported. Cases complicating type 2 HIT are also reported.


In HISN, as mentioned above, on a pathophysiologic level heparin-induced anti-platelet IgG antibodies are formed against a complex of heparin bound PF 4, a component of the a granules of platelets. The resultant antigen-antibody complex adheres to and activates platelets, which leads to the formation platelet thrombi and occlusion of vessels.

As for WISN, Vitamin K is a cofactor that is required for the gamma-carboxylation of Factors II, VII, IX, and X, as well as Protein C and Protein S. Warfarin therapy decreases this action, and thus decreases the maturation of these clotting factors. Protein C levels drop rapidly in the first few days of warfarin therapy; more rapidly than the other procoagulant factors. It is an important anticoagulant and so it is thought that the relative deficiency of mature protein C is responsible for the procoagulant state that leads to necrosis. While the reasons for particular localization to the fatty areas are unclear, it has been hypothesized that factors such as local temperature or perfusion may play a role.

Systemic Implications and Complications

In HISN, clinical or serological evidence of type 2 HIT is present in many of the reported cases in which it has been looked for. This syndrome may predispose to systemic clotting. New or increased deep venous or arterial thrombosis or pulmonary embolism may occur, which may increase morbidity and mortality in these patients. It is therefore vital to recognize the syndrome early.

In WISN, the skin and subcutaneous tissues are affected. Limb gangrene has been reported in cases where WISN has complicated type 2 HIT.

Treatment Options

Treatment options are summarized in Table I.

Table I.
Medical Treatment Surgical Treatment
Heparin-induced skin necrosis
Replace heparin or LMW heparin with an alternate anticoagulant Surgical debridement of necrotic plaques
Topical occlusive dressings
Warfarin-induced skin necrosis
Replace warfarin with an alternate anticoagulant Surgical debridement of necrotic plaques
Vitamin K
Fresh frozen plasma/ Protein C concentrate
LMW, low molecular weight.

Optimal Therapeutic Approach for this Disease

In HISN, necrotic skin has been noted to heal spontaneously over days to weeks. For deeper plaques, occlusive dressings applied to the ulcers left after the necrosis sloughs may aid healing. Large deep plaques may rarely need to be debrided.

It is imperative to work the patient up for type 2 HIT using the aforementioned assays, so as to guide the choice of replacement anticoagulation. Unfractionated heparin or an alternative LMWH product should be avoided in any patient with heparin-induced antibodies as these may cross-react

Forty to 50 percent of HIT patients have been reported to have a thrombotic event when heparin is discontinued and ongoing anticoagulation may be needed for the patient’s underlying medical concerns. In these patients non-heparin preparations, such as hirudin or fondaparinux, can be used. Recently, a low rate of cross reactivity of antiplatelet antibodies has been seen with fondaparinux. Danaparoid has also been recommended as an alternative anticoagulant. However, cross-reacting antibodies may be present. Therefore, an in vitro plasma assay with the patient’s plasma and the drug looking for platelet aggregation should be performed prior to initiation of treatment.

Direct thrombin inhibitors, such as lepirudin and argatroban, have been studied in this setting and found to be effective anticoagulants. Coumadin monotherapy may decrease protein C levels and give rise to extension of necrosis in these patients. Consultation with hematology is recommended.

In WISN, warfarin needs to be discontinued promptly. Subcutaneous vitamin K has been administered in some cases, but the dose has not been stated in the literature and should be based upon the INR. Consultation with a hematologist would be warranted.

It is important to note here that reversal of the anticoagulation will be quickly obtained if anticoagulation was life saving, then other anticoagulation in an inpatient setting might be considered based upon the scenario. Fresh frozen plasma (or protein C concentrates in known cases of protein C deficiency) can be given to restore the levels of anticoagulant factors. Heparin is recommended as an alternative anticoagulant. As opposed to HISN, the necrotic plaques in WISN are deep and usually will require surgical debridement. Mastectomy or amputation may be needed in severe cases. Skin grafting is needed in many cases.

Patient Management

In the case of heparin- or LMW heparin-induced necrosis, the patient and his or her family should be made aware of this side effect and the need to avoid heparin or heparin-containing agents in the future. Necrotic wounds and the platelet count should be monitored daily. Both should recover spontaneously. For deeper, nonhealing necrotic plaques, debridement may be offered to assist with healing. This is usually not necessary.

The risks of systemic thrombosis should be discussed with the patient and the primary team. A discussion with the consulting cardiologist or a hematologist should be held regarding the best choice for an alternative anticoagulant, based on the patient’s underlying problem, and whether heparin has been used to prophylax against thrombosis or treat it. Patients may need to be monitored in the intensive care unit setting, depending on their overall state of health.

In cases of WISN, the patient and his or her family should be made aware of this side effect. A consultation with the surgical or plastic surgical team should be requested when the patient is stable, and when there is no longer progression of necrosis. Consultation with hematology should be requested to recommend treatment of the condition, as well as to provide input on alternative anticoagulation. As with HISN, patients may be need to be monitored in the intensive care unit.

The patient should be counseled to undergo a thrombophilic work-up after the acute event, including protein C and S level screening.

Unusual Clinical Scenarios to Consider in Patient Management

In patients with skin necrosis who have received both heparin and warfarin, it may be difficult to decide which one is the culprit. As stated above, warfarin-induced necrosis typically favors sites with substantial subcutaneous fat, including thighs, buttocks, and breasts. However, warfarin-induced skin necrosis has been reported to occur in the setting of type 2 HIT and so, in atypical cases, it may be necessary to discontinue both agents, to treat with one of the direct thrombin inhibitors or heparinoids above, and to administer vitamin K and fresh frozen plasma.

What is the Evidence?

Gurbuz, AT, Elliott, WG, Zia, AA. “Heparin-induced thrombocytopenia in the cardiovascular patient: diagnostic and treatment guidelines”. Eur J Cardiothorac Surg. vol. 27. 2005. pp. 138-49. (An excellent review of heparin-induced thrombocytopenia, including pathophysiology of the syndrome, complications, and treatment guidelines.)

Handschin, AE, Trentz, O, Kock, HJ. “Low molecular weight heparin-induced skin necrosis-a systematic review”. Langenbecks Arch Surg. vol. 390. 2005. pp. 249-54. (An excellent comprehensive review of the reported cases of LMW heparin-induced skin necrosis.)

Nazarian, RM, Van Cott, EM, Zembowicz, A, Duncan, LM. “Warfarin-induced skin necrosis”. J Am Acad Dermatol. vol. 61. 2009. pp. 325-32. (An outstanding review of the clinical and histopathologic features of WISN and it's mimickers.)

Roujeau, JC, Stern, RS. “Severe cutaneous reactions to drugs”. N Engl J Med. vol. 331. 1994. pp. 1272-85. (An outstanding review of severe drug reactions, including WISN.)

Warkentin, TE. “Think of HIT”. Hematology Am Soc Hematol Educ Program.. 2006. pp. 408-14. (An outstanding review of the clinical presentations of HIT and its management by an expert in the field.)

Wütschert, R, Piletta, P, Bounameaux, H. “Adverse skin reactions to low molecular weight heparins: frequency, management and prevention”. Drug Saf. vol. 20. 1999. pp. 515-25. (An excellent review of the skin side effects of LMW heparins.)

Yombi, JC, Belkhir, L, De Baere, T, Dubuc, JE, Hainaut, P. “Low-molecular-weight heparin-induced skin necrosis: about 2 cases”. Acta Clin Belg. vol. 64. 2009. pp. 228-30. (Two further cases of LMW heparin-induced skin necrosis are described along with a review of the current understanding of pathogenesis.)