INTRODUCTION: Many pregnant women are treated with antibiotics during labor to prevent transmission of group B streptococcus (GBS, Streptococcus agalactiae) to their baby during passage of the birth canal, and so reduce the risk of serious infection of the newborn. Methods for intrapartum testing for GBS have been introduced to select women to whom intrapartum antibiotic prophylaxis should be offered. For such an intrapartum test to be useful in clinical practice, it has to be specific as well as sensitive. The aim of the present study is to evaluate the accuracy of the polymerase chain reaction (PCR) assay compared with an optimized culture method for GBS. MATERIAL AND METHODS: In the period from 12 May 2015 to 18 December 2015 we collected rectovaginal swabs from 106 women in the labor ward presenting in labor between gestational week 35+0 and 36+6 or presenting with prelabor/preterm prelabor rupture of membranes (PROM/PPROM) for > 14 h after gestational week 34+0 . We performed GBS culture (reference standard) and a molecular GBS test (Xpert GBS, Cepheid Ltd., Sunnyvale, CA, USA). RESULTS: Based on intrapartum culture, 23.6% (25/106) were colonized with GBS. Intrapartum PCR showed a colonization rate of 25.7% (27/105). The sensitivity of the test was 100% (86.28-100%). The specificity of the test was 97.5% (91.26-99.70%). The positive predictive value was 92.6%. In one case, we had no result with PCR testing, giving an invalid test rate of < 1%. CONCLUSION: The PCR test has sufficient accuracy to direct intrapartum antibiotic prophylaxis for GBS transmission during delivery.