Are You Confident of the Diagnosis?

Erythema multiforme (EM) is an acute, immune-mediated, self-limited mucocutaneous condition characterized by distinctive target lesions with concentric color variations. Erythema multiforme minor describes EM without mucosal involvement; erythema multiforme major refers to EM with mucosal disease. It should be noted that EM with mucous membrane involvement (EM major) and Stevens-Johnson syndrome (SJS) are different diseases with distinct causes. Less common variants of EM include:

–Recurrent EM: subset of patients with EM that experience frequent episodes over many years leading to substantial morbidity. Studies have found rates of recurrence of six episodes per year, with mean durations of disease ranging from 6 to 10 years.

–Persistent EM: Rare variant characterized by the uninterrupted occurrence of typical and atypical EM lesions (often widespread and papulonecrotic or bullous). May continue for longer than 1 year without therapy.

Continue Reading

  • What you should be alert for in the history

EM is frequently associated with infections, particularly herpes simplex virus (HSV) and less commonly Mycoplasma pneumoniae. Therefore it is important to inquire about signs or symptoms of HSV and cough or respiratory symptoms, which may be present in cases due to M pneumoniae infection.

Drugs can cause EM, and a thorough review of any new or recent medications should be undertaken.

Mucous membrane involvement in EM may include oral, ocular, and/or genital mucosa. Severe oral involvement may make eating and drinking particularly painful and difficult. Prodromal symptoms (eg, fever, malaise, myalgias) are uncommon in mild cases of EM but can be seen in cases with significant mucosal involvement (EM major).

  • Characteristic findings on physical examination

Cutaneous features

Clinical manifestations of EM are varied, as the term “multiforme” implies. Most lesions appear similar in a particular patient at a given time, though the morphology of lesions may vary between patients and also may evolve over the course of the disease in a single patient.

Target lesions are the hallmark of EM. Initial lesions may be round, erythematous papules that evolve into classic target lesions. Typical target lesions consist of three components: a dusky central area or blister, a dark red inflammatory zone surrounded by a pale ring of edema, and an erythematous halo on the extreme periphery of the lesion (Figure 1). Atypical target lesions may occur and manifest as raised, edematous, palpable lesions with only two zones of color change and/or a poorly defined border.

Figure 1.

Classical acral targetoid lesions of erythema multiforme. Symmetrically distributed, discrete and coalescing, targetoid papules on the hands. Note the presence of typical target lesions (three zones of color change) and atypical target lesions (two zones of color change).

EM lesions are commonly symmetric and acrally distributed, and often progress in a centripetal manner. The hands and extensor extremities are commonly involved, although the face, neck, palms, soles, flexural extremities, and trunk may also be involved. Target lesions usually involve the extremities. Lesions can be photodistributed and may occur at sites of trauma and sunburn (isomorphic [Koebner] phenomenon).

Other features sometimes observed include erythema and swelling of the nail folds and clusters of lesions on and around the elbows and knees.

Lesions are commonly asymptomatic, although itching and burning may be present. Typically EM lesions appear over 3 to 5 days and resolve in approximately 2 weeks.

Mucous membrane involvement

Painful erosions can involve the oral, ocular, and genital mucosa.

Oral involvement is most common and can affect up to 70% of patients with recurrent EM. Oral lesions include diffuse areas of mucosal erythema, bullae and erosions, and ulceration. The vermillion lip and mucosal surfaces including the buccal mucosa, labial mucosa, non-attached gingivae, and tongue tend to be affected. Mucous membrane involvement usually occurs in association with cutaneous lesions, although EM can present with mucous membrane involvement alone (usually oral). Rarely, involvement can extend to the pharynx and upper respiratory tract.

  • Expected results of diagnostic studies

The patient’s history and clinical findings provide the most important information in making the diagnosis of EM. Other studies listed below may be helpful in confirming the diagnosis of EM in selected patients.


Not specific for or diagnostic of EM. Severe cases may demonstrate elevations of the erythrocyte sedimentation rate, white blood cell count, and liver enzymes. Abnormal chest X-ray findings may be present in cases due to M pneumoniae. Some patients with persistent EM may have hypocomplementemia and/or circulating immune complexes.


Skin biopsy can help to corroborate the diagnosis of EM when it is uncertain and to exclude other disorders that may present with clinical features that resemble EM. Typical findings include basal cell vacuolar degeneration, scattered necrotic keratinocytes, lymphocyte exocytosis, and a dense superficial dermal lymphohistiocytic infiltrate around blood vessels and along the dermoepidermal junction. Early lesions may show prominent dermal changes with papillary dermal edema, lymphohistiocytic infiltrate, and red blood cell extravasation. Some lesions may demonstrate subepidermal clefting or vesiculation (due to extensive basal cell vacuolar degeneration).

Histopathologic findings are influenced by the site of the biopsy within a lesion: the dusky central portion may reveal subepidermal separation with necrotic keratinocytes, while the peripheral portion often shows more prominent dermal changes such as papillary dermal edema, vascular dilation, and a perivascular mononuclear cell infiltrate.

Mucous membrane lesions show similar histologic findings to cutaneous lesions, with more prominent spongiosis and intracellular edema leading to intraepithelial blisters.

Direct immunofluorescence microscopy findings are usually non-specific. One study found granular deposits of C3 and IgM at the dermoepidermal junction and around superficial dermal blood vessels, while fibrin was observed around dermal blood vessels and in a bandlike distribution at the dermoepidermal junction in regions of necrosis or blister formation.



Tzanck smear, polymerase chain reaction (PCR) of swab culture, viral culture, or direct fluorescent antibody testing can be used to confirm the presence of the virus in lesions suspicious for active HSV infection.

Serologic testing can exclude HSV-associated EM when negative, although antibody titers are not useful for detecting episodes of recurrent disease.

Molecular testing (PCR or in-situ hybridization) for HSV can be performed on skin biopsy specimens to confirm a diagnosis of HSV-associated EM.

M pneumoniae

Serologic testing can help confirm a diagnosis of M pneumoniae-related EM. Antibody titers rise 7 to 9 days after infection and peak at 3 to 4 weeks; thus either a single elevated immunoglobulin M antibody 7 days into the illness or a fourfold or greater increase in the immunoglobulin G antibodies in paired sera tested 2 to 3 weeks apart can confirm the diagnosis.

  • Diagnosis confirmation

Several entities may present with mucocutaneous findings that resemble EM. Moreover, typical target lesions may not always be present in each case of EM. Skin biopsy often can assist in excluding these entities. The differential diagnosis of EM and distinguishing features are enumerated below.

Urticaria: Lesions have a normal central zone, are transient (last less than 24 hours), and new lesions appear daily; in contrast to EM, in which lesions often have a central damaged zone (dusky, bullous, or crusted), are fixed for at least 7 days, and all lesions appear within the first 72 hours of the disorder.

Stevens-Johnson syndrome (SJS): Both EM and SJS may present with mucous membrane involvement and atypical target lesions on the skin. However, the atypical target lesions in SJS are flat (macular), rather than the raised (papular) lesions seen in EM. SJS is most commonly due to drugs, while EM is most commonly due to infection.

Fixed drug eruption: Generalized fixed drug eruption may be difficult to differentiate from EM both clinically and histologically, although fixed drug eruption typically has fewer lesions present at the time of first outbreak. A thorough medication history may also assist with the diagnosis.

Bullous pemphigoid: May present with urticarial lesions and/or tense bullae, with or without mucosal involvement. However, typical target lesions and dusky erythema are rare in bullous pemphigoid. Perilesional biopsy for direct immunofluorescence microscopy, indirect immunofluorescence studies, and enzyme-linked immunosorbent assay (ELISA) testing for bullous pemphigoid antigens 180 and 230 can help to confirm the diagnosis of bullous pemphigoid.

Paraneoplastic pemphigus: May present with polymorphous skin lesions (including lesions reminiscent of EM) and mucosal erosions. Routine histopathology, direct immunofluorescence microscopy, and indirect immunofluorescence can aid in differentiating this from EM.

Sweet syndrome (acute febrile neutrophilic dermatosis): Routine histopathology demonstrates a predominantly neutrophilic infiltrate that helps to distinguish it from EM.

Rowell syndrome: Characterized by the occurrence of EM-like lesions in patients with cutaneous lesions of lupus erythematosus. Patients often have a positive antinuclear antibody (speckled pattern). Histopathologic and serologic findings help to differentiate this condition from EM.

Polymorphous light eruption: Pediatric cases of HSV-induced EM that resemble polymorphous light eruption and juvenile spring eruption have been reported. A history of herpetic lesions preceding the eruption help to corroborate the diagnosis of EM.

The differential diagnosis of patients with oral EM includes paraneoplastic pemphigus, pemphigus vulgaris, mucous membrane pemphigoid, oral lichen planus, and complex aphthosis. These disorders should be considered in patients with frequently recurring disease. Skin biopsy (sent for routine histopathology and direct immunofluorescence microscopy), indirect immunofluorescence studies, and ELISA testing for desmogleins 1 and 3 and bullous pemphigoid antigens 180 and 230 are of critical importance in establishing the diagnosis.

Who is at Risk for Developing this Disease?

The annual incidence of EM is unknown; estimates have ranged from greater than 0.01% to far less than 1%. EM occurs most frequently in young adults between the ages of 20 and 40. Children and older adults can also be affected. EM exhibits a slight male predominance.

EM typically occurs sporadically throughout the year.

Genetic susceptibility may play a role in the development of EM. HLA-DQB1*0301 has been detected more frequently in EM patients than in controls, particularly in those with herpes-associated EM. Other reported HLA associations include types Aw33, DRw53, B15(62), B35, and DQ3 (although the relationship of Aw33 and DQ3 to EM has been questioned).

What is the Cause of the Disease?

  • Etiology


Infections (viral, bacterial, fungal) account for approximately 90% of cases, with HSV as the most commonly identified precipitant. The possibility of HSV-induced disease should be considered in all patients. In HSV-associated EM, the eruption usually occurs 2 to 17 days after an episode of HSV (mean interval is 8 days).

M pneumoniae is also an important cause of EM, particularly in children. A variety of other infections have been infrequently implicated, including histoplasmosis, vaccinia, orf, Epstein-Barr virus, yersinia, varicella zoster virus, hepatitis B and C viruses, tuberculosis, and parvovirus B19.

Drugs induce EM in less than 10% of cases, with the most common precipitators being non-steroidal anti-inflammatory drugs (NSAIDs), sulfonamides, antiepileptics, and antibiotics.

Other etiologic associations include malignancy, autoimmune connective tissue disease, immunizations, radiation, inflammatory bowel disease, sarcoidosis, and menstruation.

Recurrent EM

EM can be induced by asymptomatic subclinical recurrences of HSV infection. The most common cause of recurrent EM is HSV, with some studies estimating 61% to 100% of recurrent EM is due to HSV. A recent study of recurrent EM, however, found an association with HSV in only 23% of cases.

Other reported causes of recurrent EM include recurrent M pneumoniae infections, hepatitis C virus, vulvovaginal candidiasis, complex aphthosis, menses, polymorphic light eruption, and ingestion of benzoic acid, a preservative in food.

The cause of recurrent EM may be unclear in up to 60% of patients. However, it has been suggested that some cases of idiopathic recurrent EM may be due to subclinical HSV infection: studies utilizing PCR on lesional skin biopsies from patients with idiopathic recurrent EM demonstrated the presence of HSV DNA within some of the specimens.

Persistent EM

Reported in association with viral infections such as HSV, EBV, hepatitis C, and influenza; as well as inflammatory bowel disease and malignancy.

  • Pathophysiology

Much of the data exploring the pathogenesis of EM has been derived from studies investigating HSV-associated EM. The presence of HSV DNA has been demonstrated in skin biopsy specimens from patients with EM using polymerase chain reaction (PCR) testing. The development of EM secondary to HSV infection is postulated to involve a cell-mediated immune process directed against viral antigens deposited in lesional skin. A proposed mechanism for the development of EM lesions in the setting of HSV infection is as follows:

  • Virus released into the blood during reactivation of HSV infection is phagocytosed by circulating peripheral blood mononuclear cells, particularly CD34+ Langerhans cell precursors.

  • Langerhans cells containing HSV travel to the epidermis, where they transfer viral DNA fragments (including those that encode viral DNA polymerase) to epidermal keratinocytes. The migration of Langerhans cells to the epidermis is facilitated by virus-induced upregulation of E-cadherin expression on Langerhans cells and the presence of adhesion molecules (ICAM-1) on microvascular endothelial cells.

  • Expression of HSV genes encoded within DNA fragments deposited on the skin leads to recruitment of HSV-specific CD4+ Th1 cells that produce interferon (IFN) gamma in response to viral antigens.

  • The release of IFN-gamma initiates an inflammatory cascade that promotes the lysis of HSV-infected keratinocytes and the recruitment of autoreactive T-cells, leading to the epidermal damage and the inflammatory infiltrate that characterize cutaneous lesions of EM.

It is unclear whether EM related to other inciting agents follows a similar pathway. In drug-induced EM, the presence of tumor necrosis factor (TNF) alpha in lesional skin rather than IFN-gamma correlates with the development of skin lesions. It should be noted, however, that EM has been reported to occur in the setting of TNF-alpha inhibitor therapy.

It is unknown why EM does not occur in most people infected with HSV, and why it does not appear with every recurrence of HSV infection among those with a history of HSV-associated EM. Factors that may determine whether a particular patient develops EM after HSV infection include:

  • Presence of the specific peripheral blood mononuclear cell subset involved in DNA transport

  • Differences in the processing of viral DNA by phagocytic cells

  • Variation in the specific viral proteins expressed in the skin

  • Presence of epitopes shared with cellular proteins that affect the development of an autoreactive immune response

Systemic Implications and Complications

EM is typically self-limiting and completely resolves in less than 4 weeks without significant sequelae. The most common cutaneous sequelae of EM is post-inflammatory hyperpigmentation, which may persist for months after resolution of EM and may be particularly troublesome in patients with dark skin.

In severe cases of EM, elevations of the erythrocyte sedimentation rate, white blood cell count, and liver enzymes may be detected. Fluid and electrolyte abnormalities should be assessed in those with compromised oral intake.

Infrequently, ocular involvement can lead to keratitis, conjunctival scarring, and visual impairment. Esophagitis and upper airway involvement leading to pneumonia have rarely been described.

Malignancy-associated EM is rare: most commonly this has been described in the setting of underlying hematologic malignancies (ie, leukemia, lymphoma), although cases of EM (usually persistent EM or atypical EM that is unresponsive to treatment) in association with solid-organ malignancies have been reported (gastric adenocarcinoma, renal cell carcinoma, extrahepatic cholangiocarcinoma, and uterine chorionepithelioma).

In cases of recurrent EM without a clearly associated etiology, and in cases of persistent EM: it may be prudent to perform a complete history, review of systems, physical examination, and selected laboratory testing to rule out an underlying infectious, inflammatory, autoimmune, or malignant disorder.

Treatment Options


    In general, if EM is thought to be due to an active infection, appropriate treatment for the infection can be implemented. In cases due to HSV, it is generally felt that treatment with oral antivirals after the appearance of HSV-associated EM does not affect the clinical course of the eruption.

    If EM is thought to be due to a new drug, the drug should be discontinued.


    Topical corticosteroids, swish-and-spit oral solutions containing equal parts viscous lidocaine, diphenhydramine, and antacids, antihistamines


    Oral prednisone 40 to 60mg daily tapered over 2 to 4 weeks; hospitalization for nutrition and pain control in those without sufficient oral intake.


    Referral to an ophthalmologist


    –Elimination of possible inciting agent (if present)

    –Antiviral therapy


    For those with HSV-associated recurrent EM or idiopathic recurrent EM who have 6 or more recurrences per year or fewer but debilitating episodes of EM

    Oral (continuous)

    –Acyclovir 400mg twice daily

    –Valacyclovir 500mg twice daily

    –Famciclovir 250mg twice daily

    NOTE: Doses can be doubled for those who fail initial treatment

    Oral (intermittent)

    Short course (5 days) of acyclovir at initial onset of recurrent HSV infection if there exists a clearly defined association and interval between HSV infection and the onset of EM.


    Acyclovir 5% cream applied daily to areas of recurrent HSV infection is not effective for prevention of HSV-associated recurrent EM.


    –Azathioprine 100 to 150mg/day (2mg/kg/day in those with normal thiopurine methyltransferase [TPMT] activity)

    –Mycophenolate mofetil 1000mg twice daily

    –Dapsone 100 to 200mg daily

    Other systemic therapies


    –Intramuscular immunoglobulin G




    –Interferon alfa (in cases secondary to hepatitis C virus infection)

Optimal Therapeutic Approach for this Disease


In those with mild disease, symptomatic measures to reduce pain or pruritus should be the focus of treatment. Systemic glucocorticoid therapy should only be considered in those with severe oral mucosal involvement.

Potential underlying inciting agents for EM should be sought (eg, infection, drug). In general, if EM is thought to be due to an active infection, appropriate treatment for the infection can be implemented. In cases due to HSV, it is generally felt that treatment with oral antivirals after the appearance of HSV-associated EM does not affect the clinical course of the eruption. If EM is thought to be due to a new drug, the drug should be discontinued.

For those with only cutaneous involvement, topical corticosteroids can be used (medium potency for trunk and extremities; low potency for facial or intertriginous areas). Oral antihistamines can be used for patients who note itching or burning of lesions.

Painful oral erosions can be treated with a high-potency topical corticosteroid gel (eg, fluocinonide 0.05% gel applied 2 to 3 times per day) and a mouthwash containing equal parts viscous lidocaine 2%, diphenhydramine (12.5 mg/5 mL), and an aluminum hydroxide and magnesium hydroxide mixture (Maalox®) as a swish-and-spit up to 4 times per day as needed.

Severe oral mucosal involvement may result in debilitating pain, leading to an inability to ingest foods or liquids. Some patients may require hospitalization for nutrition and pain control. Oral glucocorticoids have been used to treat patients with severe mucosal involvement, although there is concern that glucocorticoids only partially suppress disease activity and may increase the risk for disease chronicity and prolonged duration of attacks. We suggest short courses of oral glucocorticoids ONLY be used in patients with severe and debilitating EM with mucosal involvement. In such instances, prednisone 40 to 60mg/day (or its equivalent) can be tapered over 2 to 4 weeks.

Patients with ocular involvement should be referred immediately to an ophthalmologist to prevent long-term sequelae.


General principles

Inciting causes of recurrent EM should be sought and eliminated when possible, although an underlying cause of recurrent EM is never found in some patients with recurrent EM despite extensive investigations. Systemic prophylactic therapy should be considered in patients who experience six or more recurrences per year or who have fewer, but debilitating, episodes of EM.

Systemic antiviral medication is the first-line treatment for HSV-induced recurrent EM and recurrent EM without an identifiable cause.

Second-line systemic agents (ie, immunosuppressive or immunomodulatory drugs) can be used in patients who do not improve with antiviral drugs.

Given that patients with recurrent EM may experience recurrences over many years, therapy to prevent recurrent disease may be necessary for a prolonged period.

Antiviral therapy

In general, continuous (daily) oral antiviral therapy is the preferred treatment of recurrent EM. In contrast, a short course (5 days) of oral acyclovir at the initial sign of recurrent HSV infection can be considered if there exists a clearly defined association and interval between HSV infection and the onset of EM. Topical antivirals are not effective for the prevention of HSV-associated recurrent EM.

In those with known HSV-associated EM, continuous antiviral therapy is very effective in inducing complete or partial responses. In those with EM that is not clearly associated with HSV, antiviral therapy is less effective, although the presence of subclinical HSV infection may account for patients who respond to therapy but have no history of cutaneous signs of HSV infection.

We suggest a 6-month trial of antiviral suppressive therapy in all patients with idiopathic recurrent EM, given the tolerability of systemic antiviral medications and the possibility of subclinical HSV infection as a cause for EM. Antiviral regimens that can be considered are:

  • Acyclovir 400mg twice daily

  • Valacyclovir 500mg twice daily

  • Famciclovir 250mg twice daily

For patients who fail initial treatment with an antiviral agent, the above doses can be doubled. If this is not effective, a trial of an alternative antiviral should be considered. Some studies have suggested valacyclovir may lead to better responses than either acyclovir or famciclovir, although the evidence is not sufficient to determine the superiority of one antiviral over another.

Second-line systemic therapies

Immunosuppressive or immunomodulatory agents should be considered in patients with recurrent EM despite a course of continuous antiviral therapy. Treatment should be continued for at least 6 months before concluding that a drug is ineffective. We suggest considering one of the following:

Azathioprine – Target dose is 100 to 150mg/day or 2mg/kg/day in patients with normal thiopurine methyltransferase [TPMT] activity. One study of recurrent EM demonstrated complete disease suppression in 11 patients with severe disease who had failed other therapies, although the condition recurred upon discontinuing therapy. Another study of recurrent EM showed less impressive results, with only two out of five patients achieving complete or partial response. TPMT (an enzyme involved in the metabolism of azathioprine) activity should be tested prior to initiation of this medication.

Mycophenolate mofetil – Target dose is 1000mg twice daily. One series of recurrent EM patients seen at a tertiary referral center found that six of eight patients treated with mycophenolate mofetil achieved complete or partial suppression of disease, although given the small number of patients, further studies are needed to determine whether mycophenolate mofetil should be the preferred treatment of severe recurrent EM.

Dapsone – Target dose is 100 to 200mg/day. Results in the literature have been mixed, with one study showing either partial or complete suppression of disease in eight of nine treated patients, while another study found that only five of ten patients had either complete or partial disease suppression. Glucose-6-phosphate dehydrogenase (G-6-PD) levels should be checked prior to initiation of dapsone.

Other systemic therapies

Although infrequently utilized for the treatment of recurrent EM, other therapies that have been reported to be beneficial in selected cases include:

  • Hydroxychloroquine

  • Intramuscular immunoglobulin G

  • Cyclosporine

  • Thalidomide

  • Cimetidine

  • Interferon alfa (in cases secondary to hepatitis C virus infection)

Patient Management

General management points

Most patients with EM can be managed with symptomatic therapy alone. In general, the treatment of EM varies according to the severity of the acute eruption and the presence or absence of recurrent disease.

Special management points to consider for patients with recurrent EM

It is important for patients and their families to understand that recurrent EM often runs a protracted course over many years, and therefore treatment to prevent recurrent disease may be necessary for a prolonged period (possibly several years). For patients that fail a 6-month trial of suppressive antiviral therapy, the recommended dose of the antiviral agent can be doubled. If this is not effective, a trial of an alternative antiviral can be considered.

Although the literature does not have sufficient data to support specific recommendations for duration of treatment, for those with recurrent EM who respond to continuous antiviral therapy, we typically treat for 1 to 2 years prior to attempting to discontinue therapy. If EM recurs after treatment is stopped, we restart the antiviral drug at the lowest effective dose and reattempt cessation therapy every 6 to 12 months thereafter. Immunosuppressive or immunomodulatory medications should be considered in patients with severe, recurrent EM who fail to respond to continuous systemic antiviral therapy.

For those receiving treatment with second-line systemic therapies (eg, azathioprine, mycophenolate mofetil, or dapsone), we recommend that treatment should be continued for at least 6 months before concluding that it is ineffective. For patients who achieve remission of their recurrent EM with one of these agents, we continue therapy for 6 to 12 months followed by a taper of the medication (over 2 to 4 months) to the lowest effective dose or to cessation. If EM recurs, we restart the drug at the lowest effective dose for 4 to 6 months prior to another attempt at tapering. Appropriate laboratory studies (eg, complete blood count, liver function tests, creatinine) should be monitored at regular intervals.

Unusual Clinical Scenarios to Consider in Patient Management

Despite optimal treatment and management, some patients with recurrent EM may have a less favorable prognosis. The following features have been found to portend a protracted clinical course and poorer prognosis:

  • Inability to identify a specific cause

  • Lack of improvement with continuous antiviral therapy

  • Severe oral involvement

  • Almost continuous use of glucocorticoid therapy for more than 1 year

  • History of use of two or more immunosuppressive agents

What is the Evidence?

Wetter, DA, Davis, MDP. “Recurrent erythema multiforme: clinical characteristics, etiologic associations, and treatment in a series of 48 patients at Mayo Clinic, 2000 to 2007”. J Am Acad Dermatol. vol. 62. 2010. pp. 45-53. (Retrospective review of 48 patients with recurrent erythema multiforme [defined as occurring on at least three occasions] seen at a tertiary referral center. Mean duration of disease was 6 years, with a mean number of episodes per year of 6.2. More than half of patients (58%) did not have an identifiable cause for recurrent EM, and HSV was found less frequently than previously reported [caused recurrent EM in 23% of patients]. Response to systemic treatments, including continuous antivirals and immunosuppressants, was varied and often suboptimal [16 of 33 patients receiving continuous antiviral treatment had either partial or complete disease suppression].Six of eight patients treated with mycophenolate mofetil had either a complete or partial response. Features of recalcitrant cases included clinicians' inability to identify a specific cause, lack of improvement with continuous antiviral therapy, severe oral involvement, extensive corticosteroid therapy, and immunosuppressive therapy with two or more agents.)

Sokumbi, O, Wetter, DA. “Clinical features, diagnosis, and treatment of erythema multiforme: A review for the practicing dermatologist”. Int J Dermatol. 2011. (A comprehensive review of the pathogenesis, clinical features, diagnosis, and treatment of erythema multiforme. Etiologic factors, clinical manifestations, diagnostic evaluation, and differential diagnostic considerations of EM are presented. Important considerations in the treatment and management of patients with acute EM and recurrent EM are discussed.)

Schofield, JK, Tatnall, FM, Leigh, IM. “Recurrent erythema multiforme: clinical features and treatment in a large series of patients”. Br J Dermatol. vol. 128. 1993. pp. 542-5. (Retrospective review of 65 patients with recurrent erythema multiforme with a mean duration of disease of 9.5 years and a mean number of annual recurrences of six. Seventy-one percent of patients had recurrent EM due to HSV and 69% of patients had oral involvement. Fifteen of 37 patients had a complete response with continuous acyclovir. Azathioprine led to complete response in all 11 patients with severe disease, but the condition recurred on discontinuing therapy.)

Huff, JC, Weston, WL, Tonnesen, MG. “Erythema multiforme: a critical review of characteristics, diagnostic criteria, and causes”. J Am Acad Dermatol. vol. 8. 1983. pp. 763-75. (A comprehensive review of EM that provides information on the clinical characteristics, diagnosis, and etiologic associations of EM. The article provides a detailed discussion of herpes-associated EM, mycoplasma-associated EM, and drug-associated EM. A historical perspective on EM is also elucidated.)

Ayangco, L, Rogers, RS. “Oral manifestations of erythema multiforme”. Dermatol Clin. vol. 21. 2003. pp. 195-205. (A review of the oral manifestations of erythema multiforme with representative clinical photographs is provided.)

Bastuji-Garin, S, Rzany, B, Stern, RS, Shear, NH, Naldi, L, Roujeau, JC. “Clinical classification of cases of toxic epidermal necrolysis, Stevens-Johnson syndrome, and erythema multiforme”. Arch Dermatol. vol. 129. 1993. pp. 92-6. (Provides an important classification scheme (based on the morphology, pattern, and distribution of cutaneous lesions, rather than on the extent of mucous membrane involvement) to help differentiate EM from SJS and TEN. By this classification scheme, EM is a separate entity from SJS; and EM major (EM with mucous membrane involvement) is NOT to be confused with SJS).

Drago, F, Parodi, A, Rebora, A. “Persistent erythema multiforme: report of two new cases and review of literature”. J Am Acad Dermatol. vol. 33. 1995. pp. 366-9. (Describes the features of persistent EM including the uninterrupted occurrence of both typical and atypical lesions, and the frequent presence of widespread and papulonecrotic or bullous lesions. Etiologic associations of persistent EM are discussed.)

Aurelian, L, Ono, F, Burnett, J. “Herpes simplex virus (HSV)-associated erythema multiforme (HAEM): a viral disease with an autoimmune component”. Dermatol Online J. vol. 9. 2003. pp. 1(Provides a framework for conceptualizing the pathogenesis of herpes-associated EM. Proposes that viral DNA fragments are transported to the skin by peripheral blood mononuclear cells, which then triggers recruitment of autoreactive T-cells and ultimately leads to an inflammatory cascade resulting in clinical lesions of EM.)

Howland, WW, Golitz, LE, Weston, WL, Huff, JC. “Erythema multiforme: clinical, histopathologic, and immunologic study”. J Am Acad Dermatol. vol. 10. 1984. pp. 438-46. (Prospective study of 42 patients with EM. Performed a detailed histopathologic and immunologic [immunofluorescence microscopy] analysis of biopsy specimens and characterized histopathologic findings in herpes-associated EM and drug-associated EM [due to sulfa medications]. The former group displayed inflammatory changes such as spongiosis and exocytosis, and focal liquefactive degeneration of the basal cell zone of the epidermis; while the latter group showed prominent histologic necrosis of epidermal cells. Immunofluorescence microscopy findings were similar in both groups.)

Sokumbi, O, Wetter, DA. “Clinical features, diagnosis, and treatment of erythema multiforme: a review for the practicing dermatologist”. Int J Dermatol. vol. 51. 2012 Aug. pp. 889-902. (This review comprehensively discusses the clinical features, etiologic associations, diagnosis, and treatment of erythema multiforme; providing guidance to the practicing dermatologist in the evaluation and treatment of patients with erythema multiforme.)