Acute bacterial meningitis in children admitted to the Queen Elizabeth Central Hospital, Blantyre, Malawi in 1996–97


Elizabeth Molyneux Dept of Paediatrics, College of Medicine, Box 360, Blantyre, Malawi. E-mail:


To design appropriate interventions, we collected clinical and demographic data prospectively on all children aged one day to 14 years admitted with a diagnosis of bacterial meningitis (BM) from April 1st 1996 to March 31st 1997 to the Queen Elizabeth Central Hospital (QECH), Blantyre, Malawi. During the study period 267 children (2.7% of all paediatric admissions) were found to have BM; 83% were under 5 years of age, 61% under one year and 23% under one month. The most common causative organisms in the post neonatal period (n= 206) were Streptococcus pneumoniae (27%), Haemophilus influenzae type b (Hib) 21%, and Salmonella typhimurium (6%). In the neonatal group (< 1 month, n= 61) the most common causes were Streptococcus agalactiae (23%), S. typhimurium (15%), S. pneumoniae (11.5%) and other Gram negative rods (11.5%). Nineteen of 21 salmonella infections were in children under one year of age and all S. agalactiae were in infants under three months. There was delay on presentation: the average length of fever was 4.6 days, 39.5% had convulsed prior to arrival and 57% had an altered level of consciousness. An initial diagnosis of malaria had probably contributed to the delay in 22.5% (42 of 186 tested). 48% were < 80% weight for age, with 18% < 60% weight for age. The overall mortality was 40%. The outcome was worst in salmonella infections, particularly neonatal salmonella BM with a case fatality rate (CFR) of 89% (8 of 9 cases). Coma on presentation worsened prognosis (mortality 64% if Blantyre Coma Score < 3, 26% if > 3). 15% of survivors had sequelae on discharge. 20% of Hib isolates were resistant to chloramphenicol, but all salmonellae were sensitive. 5% of S. pneumoniae were resistant to penicillin and 8% to chloramphenicol. Earlier access to adequate health care and awareness of BM in a malaria-endemic area would reduce mortality and morbidity. Vaccination against Hib infection would have reduced death by 18 (17%) and prevented sequelae in 7 cases.