Buscar en
Allergologia et Immunopathologia
Toda la web
Inicio Allergologia et Immunopathologia Erythema multiforme: a review of twenty cases.
Journal Information
Vol. 26. Issue 6.
Pages 283-287 (November 1998)
More article options
Vol. 26. Issue 6.
Pages 283-287 (November 1998)
Full text access
Erythema multiforme: a review of twenty cases.
Erythema multiforme: a review of twenty cases.
M A. Martín Mateos, A. Roldán Ros
Article information
Full Text
Full Text

ALLERGOL. ET IMMUNOPATHOL., 1998;26(6):283-287


Erythema multiforme: a review of twenty cases

M. A. Martín Mateos*, A. Roldán Ros and F. Muñoz-López

*Sección de Inmunología y Alergia. Unidad Integrada de Pediatría. Hospital Clínico-Hospital de San Juan de Dios. Universidad de Barcelona.


The clinical histories of twenty children hospitalized for erythema multiforme (EM) were reviewed. Three forms of the illness were defined according to clinical criteria: minor (EMm), intermediate (EMt) and major (EMM). Previous infections were associated with the process in 30% of the cases, administration of drugs also in 30%, and a combination of the two factors in 30% of the subjects. No specific etiology was found in two cases (10%). Respiratory tract symptoms and general discomfort preceded the EMM form, and evidence of digestive organ involvement was frequent in the EMm form. Erythrocyte sedimentation rate was usually high. Postinflammatory sequelae occurred in only two cases, and one subject died in the first 24 hours.

Key words: Erythema multiforme. Toxic epidermal necrolysis. Stevens-Johnson syndrome.

Allergol et Immunopathol 1998;26:283-7.


Erythema multiforme (EM) is an acute mucocutaneous syndrome of self-limited course, with a range of etiologies and a broad spectrum of clinical presentations and severity (1). The first description was by Hebra in 1866 (2, 3). Since then, the symptomatology has been considerably broadened. Three forms of presentation can be distinguished: EMm (minor), EMM (major) and EMt (intermediate). Cases that do not meet the criteria of either EMm or EMM are considered EMt (table I). The term EMM covers the Stevens-Johnson syndrome and toxic epidermal necrolysis (Lyell''s syndrome); the latter is the more severe form of presentation, characterized by the Nikolsky sign (detachment of the epidermal layer of the skin), and can represent a mortality rate as high as 30%. The difference between these two forms of EMM lies only in the inflammatory response. The most frequently associated etiological agents, according to the literature (table II) are Herpes simplex, Mycoplasma pneumoniae, and the intake of certain drugs (2, 4-6).

Table I

The erythema multiforme classification


CourseAcute, self-limited, recurrent.Acute, self-limited.
Duration1-4 weeks.1-6 weeks.
LesionsMaculo-erythematous, lesions, blisters, target or "iris" lesions, acral lesions; symmetrical.Maculo-erythematous, lesions, blisters, detachment of  epidermis, symmetrical and disseminated.
MucosaNot affected; at most, oral mucosa.Grave; two or more mucosa affected.
SequelaePost-inflammatry hyperpigmentation.Hyperpigmentation, scars, synechiae, impaired vision.

Table II

Possible causes of erythema multiforme

1. Infections:Herpes simplex
Mycoplasma pneumoniae
Infectious mononucleosis
Yersinia enterocolitica
2. Drugs:Sulphonamides
3. Others:Immunization or desensitization maneuvers
Tumor irradiation
Lupus erythematosus

(Only well-documented causes are listed).


We performed a descriptive study of EM in the pediatrics service of the Hospital Clinic i Provincial in Barcelona (HCP).

First we identified all patients admitted in the pediatrics service and discharged with a diagnosis of EM over the past fifteen years. We used computerized data from the last two years referring to discharged patients, and in earlier cases we used the written records. All discharged patients were coded according to the WHO classification.

We then reviewed all the clinical histories of these patients, excluding those for whom the data were insufficient to corroborate diagnosis (table I). Of 28 subjects initially included, eight were omitted.

The variables analyzed were: age, sex, personal history, etiological factors, prodromes, reason for visit, physical examination, complementary examinations, treatment, complications, mean hospital stay, sequelae and number of relapses.

For statistical treatment of the data we used the SPSS statistical package.


Of the twenty patients, sixteen were male, and four female (ratio 4:1). Mean age was 6.5 years (± 2.09): EMM was diagnosed in nine patients, EMt in four, and EMm in seven (Figs. 1 and 2). There was no seasonal predominance (Fig. 3). Only two patients (10%) did not present related etiological factors in previous weeks. As regards clinical history, 30% presented drug intake, 30% infections, and 30% a combination of the two factors (table III) and (Fig. 4). Respiratory processes were the predominant type of infection. Of the drugs administered, the most frequently associated were sulphonamides, acetylsalicylic acid and penicillin G. No cases of EMm were associated with sulphonamides or penicillin G.

Figure 1.--Forms of EM.

Figure 2.--Sex.

Figure 3.--Seasonal incidence.

Figure 4.--Etiology.

Table III

Etiology of erythema multiforme

Acetylsalicylic acid (ASA)1­1
Erythromycin + sulphonamides + ASA­­1
Labial herpes simplex1­­
Oral thrush1­­
Urinary infection­1­
Barbiturates + respiratory infection­­1
Sulphonamides + ASA + respiratory infection­­1
Sulphonamides + tonsillitis­1­
Penicillin G + paratiditis­­1
Penicillin G + cefotaxime + cloxacillin + pneumonia­­1
Penicillin G + tonsillitis­­1

Forty-three percent of the cases of EMm presented a history of suspicious food intake: this feature was absent in the other forms (p = 0.037). All patients with EMM presented fever, compared with 50% of those with EMt and 43% of those with EMm (p = 0.029). Fifty-five percent of the patients with EMM presented coughing, a symptom absent in EMt and EMm (p = 0.017). Abdominal pain was more frequent in EMm (43%) and EMt (22%) than in EMM (0%). Eleven percent of the patients with EMM presented pruritus, compared with 43% in EMm and 75% in EMt. Of the EMm patients, 57% presented vomiting and 43% abdominal pain (table IV).

Table IV

Clinical manifestations of the different forms of EM

Fever32 9
Arthralgia20 2
Cough00 5
Vomiting40 2
Diarrhea10 1
Pruritus33 1
Abdominal pain31 0

Pneumonia was diagnosed in three patients with EMM, two clinically and radiologically, and one by necropsy; serological studies for influenza (1/256) were obtained only in the last case (the second sample could not be obtained). In one of the patients with EMt, maxillary sinusitis was found by x-ray. In another patient, also with EMt, leukocyturia was observed (> 100 leukocytes per field). Only in one case of EMM was a skin biopsy required to corroborate the clinical diagnosis of EM. In 35% an alteration in the liver function or hepatomegaly was found, with no significant differences between the three series. VSG was also high in 66% of the patients with EMM, 25% of those with EMt and 14% of those with EMm.

During hospitalization antibiotic treatment was administered to two EMm patients, one patients with EMt and six with EMM. All patients in whom EMt or EMM was diagnosed received parenteral corticotherapy, at doses of 1-2 mg/Kg/day, from admission until stabilization or improvement of the clinical picture; the dose was then decreased. Only two patients with EMM were transferred to the ICU; one died within the first 24 hours of respiratory complications. Mean hospital stay was 11.42 days (± 2.4) in EMm, 10.75 (± 3.3) in EMt and 20.4 (± 4.5) in EMM.

Five of the twenty patients had previously presented outbreaks of EM. EMM was diagnosed in only one of these cases; a form of EMt was diagnosed in another, and EMm in the other three (one of whom presented labial herpes). In three of these patients the current outbreak was the third recurrence. Only two EMM patients presented residual pigmentation.


This study presents the first descriptive analysis of EM. Obviously the small size of the sample introduces a number of limitations.

The diagnosis of EM was based above all on clinical data (table I) (4). Complementary explorations, such as skin biopsy, cultures, and direct immunofluorescence (2, 6-8) were used when diagnosis is doubtful.

It is occasionally difficult to identify a single etiological agent, since a vast range of factors have been reported in the literature as possible causes (2, 5, 9-13). Correct identification requires an extremely detailed clinical history.

In this study, the drugs most frequently associated with the illness were sulphonamides, penicillin G and acetylsalicylic acid. The first two were not associated with EMm.

In EMM, respiratory symptomatology predominates, with coughing and fever on the days prior to admission; in EMm digestive symptomatology is the most frequent, with vomiting and abdominal pain, coinciding with the appearance of mucocutaneous lesions. These results coincide with those of most authors, in that EMM is usually associated to a larger extent with infections of a respiratory nature or with drug intake, presenting prodromes and general discomfort.

No seasonal predominance was established, a finding that corroborates recent reports in the literature (6). In the past, the illness was believed to be more prevalent in spring and autumn (2).

Even though the role of host immunity in the pathogenesis of EM has been analyzed (14, 15), it is currently difficult to single out any one mechanism, since the results obtained in both immunological and pathological studies are controversial (6). Recently, several authors have reported a strong association between HLA-DW3 and EM (6, 16, 17), which suggest a relationship between the altered immune response and the HLA system (especially in the subjects who present recurrent episodes).

Treatment of EM depends on the causes of the illness and on the severity of the clinical picture. In general, EMm requires symptomatic treatment, only. Recurrences of EM tend to be mainly outbreaks of EMm, related to Herpes simplex, although this herpes is not always identified clinically. Bateman was the first to report the association between Herpes simple and EM, in 1813 (16, 18). Today, patients with recurrent EM are usually treated with aciclovir administered orally for a period of six to eighteen months, since it has been shown to limit recurrence (2, 4-6, 15, 19) even when there is no clinical evidence of herpes. In this study we also observed that recurrences were more frequent in EMm, and found a statistically significant relationship between fever, respiratory symptomatology and EMM. The use of corticoids has generated considerable controversy, since some authors hold that they cause further complications (4) while others recommend them (20) on the grounds that at high doses they suppress the active phase of the illness in the early stages and its complications later on. Our experience, and that of the dermatology department at our hospital, leads us to support the latter view.

Some authors have suggested treatment with cyclosporine in cases of severe, resistant EM (21). Others have obtained positive results with cimetidine used to treat recurrent EM resistant to aciclovir (22). However, there is little experience to date with these drugs.

It is essential that the patient''s family be informed of the possibility of recurrence and of the importance of follow-up at the same center to ensure appropriate study and treatment.


Se realiza un estudio retrospectivo de 20 niños diagnosticados de eritema multiforme (EM). El diagnóstico se realizó desde el punto de vista clínico, dividiéndose en las formas major (EMM), intermedia (EMt) y minor (EMm). La relación varón/hembra fue 4/1. Sólo dos casos no presentaron factores etiológicos desencadenantes (10%), los 18 restantes tuvieron como antecedentes infecciones (30%), fármacos (30%), o los dos factores a la vez (30%). Se observó el predominio de pródromos con sintomatología respiratoria y malestar general en las formas de EMM, así como sintomatología digestiva en las formas de EMm. Presentaron hepatomegalia o alteración de pruebas hepáticas un 35% de los pacientes. El dato analítico alterado con mayor frecuencia fue la VSG. Sólo hubo secuelas leves (pigmentación residual) en dos pacientes y uno presentó evolución fatal.

Palabras clave: Eritema multiforme. Necrólisis epidérmica tóxica. Síndrome de Stevens-Johnson.


1. Huff JC, Weston WL, Tonnesen MG. Erythema multiforme: a critical review of characteristics, diagnostic criteria, and causes. J Am Acad Dermatol 1983;8:763-75.

2. Huff JC, Denver MC. Aciclovir for recurrent erythema multiforme caused by Herpes simplex. J Am Acad Dermatol 1988;18:197-9.

3. Von Hebra F. On diseases of skin including the exanthema. Trans by Fagge CH. London: New Syndenhan Society; 1966;1:285-9.

4. Goldberg GN. Erythema multiforme. Controversies and recent advances. In: Callen JP, editor. Advances in dermatology. London: Year Book Medical Publishers; 1987. p.p. 73-8.

5. Herrero C, Martín E. Eritema multiforme. MTA Medicina Interna 1989;7:372-9.

6. Huff JC. Immunologic mechanism of erythema multiforme. In: Norris DA, editor. Immune mechanisms in cutaneous diseases. Marcel Dekker N. Y.; 1989. p.p. 387-402.

7. Howwland WA, Golitz LE, Weston WL, Huff JC. Erythema multiforme: clinical histopathologic and immunologic study. J Am Acad Dermatol 1984;10:438-46.

8. Orton PW, Huff JC, Tonnesen MG, Weston WL. Detection of a Herpes simplex viral antigen in skin lesions of erythema multiforme. Ann Intern Med 1984;101:48-50.

9. Bielsa I, Herrero C, Font J, Mascaró JM. Lupus erythematosus and toxic epidermal necrolysis. J Am Acad Dermatol 1987;16:1265-7.

10. Lobkowicz F, Ring J, Schwarz TF, Roggendorf M. Erythema multiforme in a patient with acute human parvovirus B19 infection. J Am Acad Dermatol 1989;20:849-50.

11. Chan HL, Stern RS, Arndt KA, Langlois J, Jick SS, et al. The incidence of erythema multiforme, Stevens-Johnson syndrome, and toxic epidermal necrolysis. A population-based study with particular reference to reactions caused by drugs among outpatients. Arch Dermatol 1990;126:43-7.

12. Commission on Acute Respiratory Disease. Association of Pneumonia with Erythema Multiforme Exudativum. Arch Intern Med 1946;78:687-710.

13. Sontheimer RD, Garibaldi RA, Krueger GG. Stevens-Johnson syndrome associated with Mycoplasma pneumoniae infections. Arch Dermatol 1978:241-4.

14. Matsuoka LY, Wortsman J, Stanley JR. Epidermal antibodies in erythema multiforme. J Am Acad Dermatol 1989;4:677-80.

15. Joly P, Autra B, Lauret P. Erythème polymorphe récidivant associé a une hypogammaglobulinémmie a expression variable. Ann Dermatol Venereol 1990;117:33-4.

16. Brice SL, Krzmien D, Weston WL, Huff JC. Detection of Herpes simplex virus DNA in cutaneous lesions of erythema multiforme. J Invest Dermatol 1989;93:183-7.

17. Kämpgen E, Burg G, Wank R. Strong association of the class II antigen HLA-DQW3 with Herpes simplex virus associated erythema multiforme. J Invest Dermatol 1988;90:245.

18. Boteman T. A practical synopsis of cutaneous diseases. Longman Z, ed. Hurst, Rees, Orme and Brown. London; 1813. p.p. 236-7.

19. Burgess JA, Johnson BD, Sommers E. Pharmacological management of recurrent oral mucosal ulceration. Drugs 1990;39:54-65.

20. Jorizzo JL. Erythema multiforme. In: Callen, Jorizzo, editors. Dermatological Signs of Internal Disease. Harcourt Brace Jovanovick: W.B. Saunders Company; 1988. p.p. 70-5.

21. Wilkel CS, McDonald CJ. Cyclosporine therapy for mullous erythema multiforme. Arch Dermatol 1990;126:397-8.

22. Kurkcuoglu N, Alli N. Cimetidine prevents recurrent erythema multiforme major resulting from Herpes simplex virus infection. J Am Acad Dermatol 1989;21:814-5.


Dra. M. A. Martín Mateos

Hospital San Juan de Dios

Sección de Inmunología y Alergia

Paseo San Juan de Dios, 2

08950 Esplugues de Llobregat


Article options
es en pt

¿Es usted profesional sanitario apto para prescribir o dispensar medicamentos?

Are you a health professional able to prescribe or dispense drugs?

Você é um profissional de saúde habilitado a prescrever ou dispensar medicamentos