Picture of the Month—Diagnosis | Dermatology | JAMA Pediatrics | JAMA Network
[Skip to Navigation]
Citations 0
Special Feature
November 2007

Picture of the Month—Diagnosis

Author Affiliations


Arch Pediatr Adolesc Med. 2007;161(11):1103. doi:10.1001/archpedi.161.11.1103

Denouement and Discussion: Blueberry Muffin Baby (Extramedullary Hematopoiesis) Due to Congenital Cytomegalovirus Infection

Nasopharyngeal, conjunctival, and urine viral cultures grew cytomegalovirus (CMV). Blood CMV pp65 antigenemia was detected at a level of 58/200 000 cells and IgG to CMV was detected. The child was treated for 6 weeks with ganciclovir, and the cutaneous findings slowly faded over 3 to 4 weeks. Initial hearing screen results were normal.

Blueberry muffin baby is a term used to describe a purplish or bluish-red maculopapular rash present at birth, typically on the face, neck, and trunk.1 Traditionally, it has been used to describe the cutaneous findings of congenital rubella, but other conditions can produce the same skin findings. The differential diagnosis of blueberry muffin baby includes conditions associated with dermal extramedullary hematopoiesis, infiltrative neoplastic lesions of the skin, and cutaneous vascular anomalies. Rarely do new lesions appear after birth. In most cases, the lesions evolve into tan macules and fade completely within a few weeks. When lesions do not fade or when they progress and enlarge, a neoplastic disorder should be suspected.

Evaluation of blueberry muffin baby begins with careful review of the pregnancy history and prenatal laboratory studies with a special focus on infectious serologies as well as blood type and antibody studies. Gross and microscopical examination of the placenta should be performed. Physical examination of the neonate should focus on growth parameters, skin and mucous membrane abnormalities, ophthalmologic findings, cardiopulmonary assessment, presence of abdominal masses or hepatosplenomegaly, and evaluation of the musculoskeletal and neurologic systems. Laboratory evaluation includes complete blood cell count with differential, platelet count, reticulocyte count, and liver function tests. Where indicated, appropriate cultures and serologies should be obtained, as should coagulation testing or imaging studies. When the diagnosis is not clear or lesions are progressive, a skin biopsy should be performed.

In this case of blueberry muffin baby, the diagnosis was congenital CMV infection, the most common cause of congenital infection worldwide. Approximately 10% of infected neonates are symptomatic at birth, with findings that may include growth retardation, jaundice, purpura or petechiae, hepatomegaly, splenomegaly, and microcephaly.2 Manifestations of central nervous system involvement are common in infected infants and include lethargy, abnormal tone, poor feeding, chorioretinitis, sensorineural hearing loss, periventricular intracranial calcifications, and other cortical abnormalities. Abnormal laboratory results include leukopenia, anemia, thrombocytopenia, hyperbilirubinemia, and elevation of transaminase levels.

The diagnosis of congenital CMV infection is established by detecting the virus by culture or other techniques in urine, blood, or other tissues obtained during the first 3 weeks of life. Many laboratories use rapid culture techniques combined with direct immunofluorescence detection with antibodies against intermediate-early or early CMV antigens.3 More sophisticated methods of CMV detection include measurement of CMV antigenemia using either indirect immunofluorescence with antibodies specific for CMV pp65 in polymorphonuclear leukocytes or CMV nucleic acid amplification techniques. Amplification strategies allow for qualitative and/or quantitative assessment of CMV RNA or DNA in clinical samples, most commonly urine or polymorphonuclear leukocytes. To our knowledge, trials studying the specificity and sensitivity of these techniques in the diagnosis of congenital CMV infection have not been performed. Serological tests, especially for the detection of IgM anti-CMV antibody at birth, may be suggestive of congenital infection, but confirmatory viral isolation and/or identification is recommended.

Numerous agents are available to treat CMV infection, including ganciclovir, its valine ester valganciclovir hydrochloride, foscarnet sodium, cidofovir, and CMV immune globulin. There is limited experience in treating congenital infection with these agents.2 However, infants with symptomatic congenital CMV infection and evidence of central nervous system involvement treated for 6 weeks with ganciclovir have less hearing loss at age 6 months and fewer developmental delays at age 12 months compared with untreated controls.4,5 Concerns of marrow suppression and other potential long-term effects such as germ cell toxic effects and carcinogenicity have led many to restrict the use of ganciclovir to the treatment of congenital CMV with central nervous system involvement until its safety and efficacy have been rigorously demonstrated.

Correspondence: Patrick G. Gallagher, MD, Department of Pediatrics, Yale University School of Medicine, 333 Cedar St, LCI 402, PO Box 208064, New Haven, CT 06520-8064 (patrick.gallagher@yale.edu).

Accepted for Publication: January 29, 2007.

Author Contributions:Study concept and design: Gallagher. Acquisition of data: Gaffin. Analysis and interpretation of data: Gaffin. Drafting of the manuscript: Gaffin. Critical revision of the manuscript for important intellectual content: Gallagher. Study supervision: Gallagher.

Financial Disclosure: None reported.

Holland  KEGalbraith  SSDrolet  BA Neonatal violaceous skin lesions: expanding the differential of the “blueberry muffin baby.”  Adv Dermatol 2005;21153- 192PubMedGoogle ScholarCrossref
Ross  SABoppana  SB Congenital cytomegalovirus infection: outcome and diagnosis.  Semin Pediatr Infect Dis 2005;16 (1) 44- 49PubMedGoogle ScholarCrossref
Schleiss  MR Antiviral therapy of congenital cytomegalovirus infection.  Semin Pediatr Infect Dis 2005;16 (1) 50- 59PubMedGoogle ScholarCrossref
Kimberlin  DWLin  CYSanchez  PJ  et al. National Institute of Allergy and Infectious Diseases Collaborative Antiviral Study Group, Effect of ganciclovir therapy on hearing in symptomatic congenital cytomegalovirus disease involving the central nervous system: a randomized, controlled trial.  J Pediatr 2003;143 (1) 16- 25PubMedGoogle ScholarCrossref
Oliver  SCloud  GSanchez  P  et al.  Effect of ganciclovir therapy on neurodevelopmental outcomes in symptomatic congenital cytomegalovirus infections involving the central nervous system: a randomized, controlled study.  E-PAS 2006;592540.2Google Scholar