LUMBAR PUNCTURE IN CHILDREN WITH PERIORBITAL AND ORBITAL CELLULITIS

Objective: To measure the frequency with which lumbar puncture (LP) was performed in children with periorbital and orbital cellulitis in the years 1979 through 1991, and to examine its value as a diagnostic test in these patients. Design: Retrospective review of both computerized discharge diagnoses and medical records. Setting: Tertiary care hospital. Patients: Two hundred fourteen children younger than 18 years of age discharged between Jan. 1, 1979, and Dec. 31, 1991, with a final diagnosis of periorbital or orbital cellulitis. Main outcome measures: Rates of LP and of bacterial meningitis. Results: Of the 214 patients, 96 (45%) had LPs. Two children (1%) had meningitis; in both, the diagnosis was strongly suspected before the LP was performed. No other subject had a discharge diagnosis of meningitis. During the 13 years of study, the proportion of patients in whom LP was performed increased from 14% in the first 6 years (1979 through 1984) to 62% from 1985 through 1991 (p < 0.0001). Those older than 2 months who were afebrile on arrival in the emergency department were 12.4 times more likely to have an LP in 1985 through 1991 than in 1979 through 1984 (p = 0.005). Conclusions: The increase in the number of LPs performed from 1985 through 1991 in children with periorbital or orbital cellulitis was not accompanied by an increase in the diagnosis of early or unsuspected meningitis. These data suggest the need for more judicious use of LP in children with periorbital swelling.