This study was performed prospectively. Between 1. January 1995 and 31 December 1997, a modification of the screening-based strategy protocol was implemented. Antenatal screening cultures for GBS were performed at approximately 30-32 weeks of gestation. The protocol recommends the use of antibiotic prophylaxis to GBS positive women with any of the obstetric risk factors for early-onset of GBS disease. Our regiment for prophylaxis for patients in labor was ampicillin 2 g. intravenously then 1 g. i.v. every 4 hours until delivery. Before this study had started (1984-1994), there were 149 serious neonatal GBS infection (149/15,040 pregnancy, among them were 97 premature infants. Thirty-one infants suffered from connatal sepsis. We observed 29 lethal infection. Between January 01, 1995, and December 31, 1997, 4150 women participated in this investigation. The incidence of positive group B Streptococcus cultures from the vaginal samples was 11.6% (481/4150). During the study period (3 years), serious GBS infection was detected in 46 infants (1.1%). There were 9 cases of neonatal sepsis (0.2%), two of them suffered lethal infections (0.05%). There was no late onset of GBS disease and lethal outcome in the last two years. Our investigations bears clinical importance because we confirmed that group B streptococcal colonization is an important risk factor for neonatal infection. The selective intrapartum chemoprophylaxis is a safe and effective intervention to prevent early-onset severe GBS disease.