Severity of childhood bacterial meningitis and duration of illness before diagnosis

Lancet. 1991 Aug 17;338(8764):406-9. doi: 10.1016/0140-6736(91)91032-p.

Abstract

Rapid diagnosis of childhood bacterial meningitis (BM) is generally believed to be essential to avoid poor outcome. To see whether duration of illness before admission to hospital was related to the severity of illness, data from children with BM diagnosed in 18 paediatric hospitals in Finland from 1984 to 1989 were collected prospectively. We divided 286 cases with culture-positive cerebrospinal fluid (CSF) into three groups: BM with a history of up to 24 h (short-history group, n = 141), of more than 24 h and up to 48 h (intermediate-history group, n = 75), and of more than 48 h (long-history group, n = 70). The longer the history, the better the clinical condition of the child. If symptoms or signs of BM lasted 48 h or less, the child did significantly worse, as judged by seven variables, than if the history was longer than 48 h (level of consciousness, p less than 0.001; seizures, p less than 0.01; CSF protein concentration, p less than 0.001; positive CSF gram-stain, p less than 0.01; positive blood culture, p less than 0.05 in Haemophilus influenzae meningitis; serum C-reactive protein, p less than 0.01 between intermediate-history and long-history groups; and urine sodium concentration, p less than 0.001). The differences were not affected by causative organism, sex, age, or preadmission oral antimicrobial agents. The findings show that if BM follows an insidious pattern of disease, diagnostic delay may be unavoidable, which may have medicolegal implications.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Age Factors
  • Analysis of Variance
  • Bacterial Infections / diagnosis
  • Bacterial Infections / physiopathology*
  • Chi-Square Distribution
  • Child
  • Child, Preschool
  • Female
  • Humans
  • Infant
  • Male
  • Meningitis / diagnosis
  • Meningitis / physiopathology*
  • Prospective Studies
  • Regression Analysis
  • Severity of Illness Index
  • Time Factors