Are You Confident of the Diagnosis?

What you should be alert for in the history

Erthema toxicum neonatorum (ETN) is an innocent, self-limited eruption that is present in approximately half of term neonates, with onset typically between 24 and 48 hours after birth, though cases have been reported from birth to the fourteenth day of life. The eruption may wax and wane over the first 2 weeks of life. Individual lesions typically last a few days, but may resolve in several hours or last longer than a week.

Characteristic findings on physical examination

Physical examination reveals a combination of pink, erythematous macules, papules and small pustules distributed over the face, trunk, and extremities (Figure 1). The eruption may involve palmar and plantar surfaces, but typically spares these areas. The number of lesions varies from a few to a few hundred. It is often referred to as having a “flea-bitten” appearance, and the background erythematous macules are often far more striking than the pustules.

Figure 1.

Erythema Toxicum neonatorum, 2 days after birth. Pink macules and papules. (Courtesy of Bryan Anderson, MD)

Continue Reading

Expected results of diagnostic studies

In typical cases, no ancillary testing is necessary, but in striking or atypical cases (based on appearance or timing of onset), a smear from a pustule examined with Wright’s or Giemsa stain will show a striking predominance of eosinophils in comparison with neutrophils. Histopathology shows concentrations of eosinophils around the pilosebaceous unit at the dermoepidermal junction. Subcorneal or superficial intradermal pustules are present. Mast cells may also be seen. Up to 15% of infants will have peripheral blood eosinophilia.

Diagnosis confirmation

Diagnosis is typically confirmed by the characteristic clinical appearance in an otherwise healthy neonate. Important and more ominous considerations in the differential diagnosis include herpes simplex, congenital cutaneous candidiasis, and impetigo.

Herpes simplex neonatorum presents in one of three forms: involvement of skin, eyes and/or mouth (SEM), central nervous system (CNS) disease, or disseminated disease presenting with CNS, hepatic, pulmonary, and hematologic involvement. Prognosis is strongly related to degree of involvement. Given that only 5% of infections occur in utero, most infants will not have clinical signs of HSV at birth. Studies show that 85% of infections occur at delivery, and approximately 10% occur from nonmaternal sources postpartum. Average incubation is 1 week, but can be as many as 3 weeks. Thus, time course can help to distinguish many cases from ETN.

Nearly 70% of infants will have vesicles. These can be sent for viral culture or direct fluorescent antibody (DFA) testing. Affected infants may have typical clustered lesions, but many neonates will present with disseminated papulovesicles, making distinction sometimes difficult. On examination, HSV lesions are often more inflammatory than those of ETN, but this is not absolute. A supportive history of genital HSV infection in the mother should be sought.

Congenital cutaneous candidiasis typically presents in the first hours of life, helping to distinguish it from ETN, with erythematous macules and development of pustules that involve the trunk, extremities, face and scalp. The oral mucosa is spared. Risk factors include premature rupture of membranes (PROM) and candidal infection of the vaginal and cervical mucosa. Infants with disseminated neonatal candidiasis show other signs of sepsis, including respiratory distress. Cultures can help to distinguish this from ETN.

Impetigo, or impetigo neonatorum, presents from day 2 to 14 of life, and is comprised by vesicles and pustules with surrounding erytema, which rupture to reveal moist denuded skin. Distinguishing factors from ETN include propensity to occur in skin folds. Bacterial culture confirms the diagnosis.

Less worrisome considerations include transient neonatal pustular melanosis distinguished by its typical presentation at birth, most commonly in darker-skinned infants, resolution of pustules over 1 to 2 days, leaving a fine collarette of scale and hyperpigmented macules that resolve over the first 1 to 3 months of life;

Milia, which are most typical on the face but occasionally involve other areas, present as 1- to 2-mm white to yellow pustules that generally resolve in the first month and require no treatment;

Miliaria includes either miliaria crystallina characterized by pinpoint superficial vesicles that rupture easily or miliaria rubra that presents with combinations of papulovesicles and pink papules in areas of occlusion, typical in infants who are slightly overheated during swaddling. Therapy for miliaria centers around cooling measures and lightweight clothing. Both types of miliaria are most common in the first months of life.

Who is at Risk for Developing this Disease?

ETN most commonly affects term newborns, typically between 24 and 48 hours after birth, almhough cases have been reported from birth to approximately 2 weeks of age. It is equally common among all races and both sexes. It is more common in neonates born at term and weighing more than 2500gm.

What is the Cause of the Disease?

The etiology of the condition remains unknown. It has at times been postulated to be allergic, infectious, or hematologic in origin. The hypothesis that it involves a graft-versus-host phenomenon against maternal lymphocytes appears to be unsubstantiated. Some authors suggest that ETN is an inflammatory response to colonization of the skin by normal bacterial flora at birth.

One Chinese study from 2005 suggests that term birth, first pregnancy, birth during summer or fall, use of formula, and vaginal delivery may predispose infants to ETN. This study also found that more extensive eruptions occured after longer duration of labor.


Upregulation of IL-1, IL-8, eotaxin, psoriasin, nitric oxide synthasess, and the aquaporins AQP1 and AQP3 have been shown to be present in ETN. Macrophage expression of the proinflammatory cytokine, high-mobility group box chromosomal protein 1 (HMGB-1) may be involved in the pathogenesis as well.

Systemic Implications and Complications

There are no systemic disorders associated with ETN.

Treatment Options

As the eruption is self-limited, no treatment is necessary aside from parental reassurance.

Optimal Therapeutic Approach for this Disease

The optimal therapy for ETN is to make the diagnosis correctly, rule out other potential causes of a similar eruption in the differential diagnosis, and provide anticipatory guidance and reassurance to the parents.

Patient Management

Infants should be managed by providing reassurance and education to the family. Routine well child check at 2 weeks of age will in most cases reveal that the majority of lesions have fully resolved.

Unusual Clinical Scenarios to Consider in Patient Management

While most common in term infants, ETN can occur in premature infants, and in such cases may have a delayed onset, making the differentiation of this from other vesiculopustular disorders in the neonatal period somewhat more difficult. In term infants as well, cases may occur as late as 2 weeks of life.

What is the Evidence?

Berg, F, Solomon, L. “Erythema toxicum Neonatorum”. Arch Dis Child. vol. 62. 1987. pp. 327-8. (This is an article from the earlier literature with a good clinical summary of ETN, including a very good section of differential diagnosis.)

Liu, C, Feng, J, Qu, R, Zhou, H, Ma, H, Niu, X, Dang, Q, Zhang, X, Tian, Z. “Epidemiologic study of the predisposing factors in erythema toxicum neonatorum”. Dermatology. vol. 210. 2005. pp. 269-72. (This is large study of 783 Chinese newborns examining potential environmental and epidemiologic factors related to onset of ETN. This study confirms earlier reports of increased frequency with term birth, but also reports a higher frequency of ETN in males, first-pregnancy births, vaginal delivery, and use of formula.)

Marchini, G, Nelson, A, Edner, J, Lonne-Rahm, S, Stavréus-Evers, A, Hultenby, K. “Erythema toxicum neonatorum is an innate immune response to commensal microbes penetrated into the skin of the newborn infant”. Pediatr Res. vol. 58. 2005. pp. 613-6. (Marchini's group has done the majority of research examining the potential pathogenesis of ETN, isolating several proteins expressed more often in cases than controls (see articles below.) This article proposes the reasonable hypothesis that ETN is part of the innate immune response as infants are colonized with bacteria soon after birth.)

Marchini, G, Hultenby, K, Nelson, A, Yektaei-Karin, E, Ståbi, B, Lonne-Rahm, S, Ulfgren, AK, Brismar, H. “Increased expression of HMGB-1 in the skin lesions of erythema toxicum”. Pediatr Dermatol. vol. 24. 2007. pp. 474-82. (See comment from Marchini's work in Pediatr Res, above.)

Marchini, G, Ståbi, B, Kankes, K, Lonne-Rahm, S, Østergaard, M, Nielsen, S. “AQP1 and AQP3, psoriasin, and nitric oxide synthases 1-3 are inflammatory mediators in erythema toxicum neonatorum”. Pediatr Dermatol. vol. 20. 2003. pp. 377-84. (See comment from Marchini's work in Pediatr Res, above.)