ADENOVIRUS CONJUCTIVITIS MIMICS PRESEPTAL AND ORBITAL CELLULITIS IN YOUNG CHILDREN

Abstract

Bacterial infections of the preseptal and orbital spaces in children typically cause marked edema and erythema of the eyelids and frequently require hospitalization for intravenous antibiotic treatment. It is not well-appreciated in the pediatric literature that adenovirus infections of the conjunctiva in young children can cause an infection very similar in appearance to a bacterial periocular infection. In older children and adults with adenovirus infections, the amount of eyelid swelling is much less and corneal findings predominate. The purpose of this study is to document how conjunctivitis caused by adenovirus in young children may be misdiagnosed as a more serious periocular bacterial infection and thereby lead to unnecessary or prolonged hospitalization for evaluation and treatment. Methods. Between January, 1991, and February, 1995, we prospectively evaluated children who had been referred by a pediatrician or ophthalmologist to the emergency room of Children's Hospital of Wisconsin or already hospitalized by their pediatricians for management of periocular infection. Results. Among 80 children with signs of preseptal or orbital infection who were seen in consultation by the Ophthalmology Service during the study period, 13 were diagnosed by us to have an ocular infection caused by adenovirus rather than by bacteria. This diagnosis was confirmed in 10 children by a positive conjunctival culture for adenovirus types 8 or 19. In the other 3 patients the results of viral cultures were negative, but the diagnosis of adenovirus conjunctivitis was considered to be very likely because of a history of exposure to someone with probable adenovirus conjunctivitis (substantiated in each case by the presence of corneal subepithelial opacities which are indicative of a recent adenoviral infection), clinical signs and disease course. In 6 of the patients there was a history of exposure to a person with a recent eye infection. The mean age of these children was 15 months with range of 5 months to 5 years. Only one patient was older than 2 years. Each patient had at least a 2-day history of a “red eye” before coming to the emergency room or being admitted to the hospital, and nine children had been treated with oral or topical antibiotics. The 11 children who were admitted were hospitalized for a mean of 3.5 days. All patients showed clinical improvement despite discontinuation of intravenous or oral antibiotics within 24 h of ophthalmologic consultation, and no child required resumption of parenteral antibiotics or readmission to the hospital. Blood cultures for bacteria were uniformly negative, and no conjunctival culture yielded a pathogenic bacterial isolate. Five children underwent orbital computed tomography scanning before ophthalmologic consultation; all 5 scans showed preseptal edema but no signs of orbital involvement. The cases were clustered in the cold weather months with 10 of the 13 occurring from December through March. No case of adenovirus conjunctivitis was seen in the summer months. The most frequent physical sign in addition to eyelid edema and erythema (Fig. 1, top) was an easily visible, whitish membrane on the palpebral conjunctiva that was occasionally hemorrhagic; this was seen in 12 patients on initial examination and in one patient on reexamination 3 days later (Fig. 1, bottom). Three patients had bilateral disease. A preauricular lymph node was palpable in four patients. Superficial keratitis was present in two patients. A conjunctival follicular response was not prominent in most patients, although many were difficult to examine with a slitlamp biomicroscope because of limited cooperation. Seven patients had fever >38°C at some time before ophthalmologic consultation; the duration of fever in hospitalized patients was 12 to 48 h. Discussion. Adenovirus conjuctivitis is associated with much more ocular adnexal inflammation in young children than in adults in whom bulbar conjunctival and corneal findings predominate. Because of the striking amount of eyelid edema and erythema that can occur in young children with adenovirus conjuctivitis, this diagnosis is often confused with more serious bacterial infections of the preseptal space or orbit. In spite of this adenovirus conjuctivitis is generally not listed in the differential diagnosis of preseptal and orbital cellulitis in children.1, 2 A lack of appreciation of age-related differences in the clinical course of ocular adenovirus infection leads to unnecessary and prolonged hospitalization for intravenous antibiotics and exposes hospital personnel and other patients to this contagious virus. Ophthalmologic examination of a child, particularly an infant or toddler, with a worrisome-appearing periocular infection might prevent unnecessary hospitalization or shorten the duration of hospitalization and amount of testing done in others. An inflammatory membrane on the palpebral conjunctiva (a pseudomembrane) was a consistent physical finding and appeared to be a relatively specific sign of infection with adenovirus. This finding should be looked for on examination. Only about one-half of patients have systemic findings such as fever or pharyngitis.3 The presence of fever, a preauricular lymph node or a follicular response of the conjunctiva was not helpful in establishing the diagnosis of viral conjuctivitis in our patients. Thorough history taking is important because it often discloses a family member or playmate with a recent ocular infection. For physicians who see children with eye infections, a rapid and inexpensive clinical test with high specificity for ocular adenovirus infection would be helpful.4, 5 A positive test result would provide support for a decision to observe the child as an outpatient with ophthalmologic evaluation whereas a negative result in a child with a worrisome-looking eye infection would substantiate the need for hospitalization. Viral cultures, which take 1 week or longer, are of confirmatory and epidemiologic value only and are not useful in either the initial evaluation of the child or the subsequent management of the infection. Acknowledgments. Supported in part by an unrestricted grant from Research to Prevent Blindness, Inc., New York, NY. Mark S. Ruttum, M.D.; Gregory Ogawa, M.D.; Eye Institute of the Medical College of Wisconsin Children's Hospital of Wisconsin Milwaukee, WIFIG. 1: Top, 7-month-old boy with severe edema and erythema of the right upper and lower eyelids; bottom, a pseudomembrane was present on the palpebral conjunctiva (arrows). Adenovirus type 8 grew in conjunctival culture.