Antibiotic prophylaxis and non-group B streptococcal neonatal sepsis

To assess the effect of increased use of intravenous penicillin for group B streptococcus (Streptococcus agalactiae, GBS) antibiotic prophylaxis on non-GBS neonatal sepsis and antibiotic resistance.

We undertook a nonconcurrent cohort study. Microbiology cultures originating from infants with early-onset neonatal sepsis in our neonatal intensive care unit (NICU) from 1992 to 1999 were reviewed. Prevalence of non-GBS neonatal sepsis in the control period (January 1, 1992, through June 30, 1995) was compared with that in the study period (October 1, 1995, through August 31, 1999), when the protocol changed. Chi-squared or Fisher exact tests were used to determine statistical significance. Resistance patterns were compared in similar fashion.

The prevalence of non-GBS neonatal sepsis was 1.2 per 1,000 (36 of 31,133) live births before and 1.1 per 1,000 (32 of 28,733) live births after institution of the Centers for Disease Control and Prevention culture-based protocol (P = .97). Our power analysis assumed a doubling in the rate of non-GBS sepsis and required 21,220 live births per arm. Gram-negative and gram-positive sepsis prevalences were not significantly different. Escherichia coli and GBS resistance patterns did not change.

Institution of a protocol for GBS antibiotic prophylaxis significantly decreased the rate of GBS neonatal sepsis but did not increase the rate of non-GBS neonatal sepsis. Antibiotic resistance patterns of these organisms were not affected.