Abstract
Objective. To explore risk factors for group B Streptococcus (GBS) colonization during pregnancy and at delivery, estimate the predictive value of early GBS colonization for colonization at delivery and in the newborn, and explore the relationship to adverse perinatal factors. Design and setting. Cohort study of pregnant women from three communities in Zimbabwe. Methods. Information collected by questionnaire at inclusion and from delivery records. Vaginal and rectal swabs collected for GBS culture at 20 and 26 weeks gestation, at delivery and from the newborn infant. Main outcome measures. GBS colonization in pregnancy, colonization of mother and newborn, and perinatal factors. Results. GBS culture results were obtained at one or more occasion for 780 (75.2%) of 1,037 women recruited. Altogether, 470/780 women (60.3%) tested positive for GBS, with colonization rates at 20, 26 weeks and delivery of 47%, 24.2% and 21%, respectively. Positive GBS culture at 20 and 26 weeks gestation had a low positive predictive value on colonization at delivery and in the newborn. Women living in rural areas were significantly more often colonized than those who lived in urban areas (p < 0.001). Other socio-economic, demographic and obstetric factors were not statistically associated with GBS colonization. GBS transmission was not statistically significantly associated with adverse perinatal outcomes. Conclusions. GBS colonization was common among pregnant women in Zimbabwe. Dwelling in a rural area was significantly associated with GBS colonization while other risk factors were not. Early GBS colonization had a low predictive value for colonization at delivery and colonization was not associated with adverse perinatal outcome.
Acknowledgements
The study was part of the study ‘Urogenital occurrence and persistence of group B streptococcal colonization in pregnancy and pregnancy outcome in Zimbabwe', a collaborative study between the Department of Medical Microbiology, College of Health Sciences, University of Harare, Zimbabwe (UZ) and the Department of Laboratory Medicine, Women and Child health care at the Norwegian University of Science and Technology (NTNU), Trondheim, Norway. The principal investigators are Professor Sylvester Rodgers Moyo (UZ) and Professor Lars Bevanger (until 2006), Associate Professor Arne Myhre from 2006 to 2008, and Associate Professor Jan Egil Afset (2008). The study was funded by a grant from NUFU Pro 42/2002.
Declaration of interest: The authors declare no conflicts of interest.