The aim of this retrospective study was to evaluate the clinical efficacy in terms of mortality and long-term morbidity of third generation cephalosporins and amikacin in combination for the treatment of gram-negative bacterial meningitis in a homogeneous group of neonates. A 15-year experience (1983–1997) with 72 term neonates without central nervous system anomalies and with gram-negative organisms grown in their cerebrospinal fluid treated with the above combination of antibiotics is presented. All isolated organisms were sensitive to cefotaxime or ceftazidime and to amikacin but 80% were resistant to ampicillin. The predominant infecting organism was Escherichia coli (68.0%) which was sensitive to both cefotaxime and amikacin in all cases but resistant to ampicillin in 48% of cases. Survival at discharge was 97.2% but ultimate survival was reduced to 94.4%, as 2 patients died a few months following discharge of conditions unrelated to meningitis. Ventriculitis was diagnosed in 10 neonates (13.8%). Among survivors, 1 neonate (1.3%) developed hydrocephalus needing shunting and 1 neonate (1.3%) with Proteus mirabilis developed a brain abscess with relapse of meningitis which was successfully treated with a 6-week course of chloramphenicol. At follow-up at an age greater than 6 months, 91.1% of the surviving infants were normal, while 92.3% of survivors at an age greater than 6 years were normal and attended normal school. These results, despite any reservations due to the nature of the study (retrospective, uncontrolled study), strongly support the use of third generation cephalosporins and amikacin in combination for the treatment of neonatal gram-negative bacterial meningitis.

In this prospective study, the 24-hour gastric aspirate volume was carefully recorded before, 24 and 48 h after administering 1.7 mg/kg/8-hourly of oral erythromycin to 16 ventilated neonates less than 32 weeks of gestation. Their median gestational age was 28.5 weeks (range 23–31 weeks), their median birthweight was 1,045 g (range 690–1,560 g) and the median day of life at which erythromycin was commenced was 9.5 days (range 4–16 days). Prior to administering erythromycin median 24-hour gastric aspirate volume, expressed as a percentage of the milk volume given over the same period, was 38.5% (range 20.0–100%). It was significantly lower 24 h (median 12%, range 0–41%, p = 0.0004) and 48 h (median 5%, range 0–21%, p = 0.0004) after commencing erythromycin. There was also significant reduction of gastric aspirate volume between 24 and 48 h after commencing erythromycin (p = 0.0024). Milk volume increment over the same period was not significant (p = 0.1022). These preliminary results warrant further evaluation through a randomised controlled trial.