Amniorrhexis lowers the incidence of positive cultures for group B streptococci Cassandra E. Henderson, MD," Howard Egre, MD," Russell Turk, MD," Victor Aning, BS,b George Szilagyi, MD,b and Michael Y. Divon, MD" Bronx, New York OBJECTIVE: Our purpose was to determine the effect of amniorrhexis on cultures for group B streptococci. STUDY DESIGN: Intrapartum vaginal cultures for group B streptococci were obtained before and after rupture of membranes in 166 women; in addition, rectal cultures were obtained from the last 35 patients. RESULTS: Beforerupture of membranes 49 of 166 (30%) of the vaginal cultures were positive for group B streptococci; only 28 of 166 (17%) remained positive after rupture of membranes (p = 0.0009). Similarly, before rupture of membranes 10 of 35 (30%) rectal cultures were positive, whereas only three of 35 (0.9%) remained positive for group B streptococci after rupture of membranes (p = 0.03). CONCLUSION: The prevalence of positive genital cultures for group B streptococci is significantly lower immediately after rupture of membranes. (AM J OBSTET GYNECOL 1993;168:624-5.) Key words: Amniotic fluid, group B streptococci Since 1978 routine culturesf or group B streptococci have been performed on all women admitted to the lab or and delivery suite of the teaching hospital of the Albert Einstein College of Medicine-Montefiore Medi- cal Center. A recurrent dil emma in thi s screened pop- ulation is the value of negative cultures after rupture of membran es. Patterson et al. I pro spectively evaluated 20 pat ients and suggested that positive endocervical cul- tu re s wer e not significantly affected by ruptured mem- br anes for up to 4 hours . However, their sample size was sma ll, and only two study patients had positive gr oup B streptococci cultures. Other investigators have suggested negative cultures may be caused by a "wash- out" or by the bacteriostatic effect of amniotic fluid ! ' 3 To evaluate the eff ect of amniotic fluid on cultures for group B streptococci we undertook this study to assess the incidence of positive cultures before and after amniotomy. Material and methods The study population included 166 women with in- tact membranes who were admitted to the labor and del ivery suite. All pat ients had vag ina l culture s for group B streptococci performed durin g the initial phys- From the Division of Maternal-Fetal Medicine, Department of Ob- stetrics and Gynecology,· and the Department of Microbiology, J ack D. Weiler Hospital of the Albert Einstein College of Medicine. Received for publication April 2, 1992; revised J uly 16, 1992; accepted July 22, 1992. Reprint requests: Cassandra E. Henderson, M D, Department of Obstetrics and Gynecology, Jack D. Weiler Hospital of the Albert Einstein College of Medicine, 1825 Eastchester Road, Bronx, N Y 10461. Copyright © 1993 Mosby-Year Book, Inc. 0002-9378/93 $1.00 + .20 6/1/41198 624 ical ex amination. The last 35 consecutive patients also had rectal cultures for group B streptococci in addit ion to the vaginal cultures. A rapid culture technique, with a sensitivity of 98 .5% and a specificity of 100%, has b een used at our institu- tion.' The use of similar rapid culture techniques ha s been reported by other inve stigators." 6 The culture medium was a selective broth medium consisting of Todd-Hewitt broth, 5% defibrinated sheep blood , 8 IJ.g/ml gentamicin sulfate, and 15 ug/ml nalidixic acid. The blood agar plate used trypticase soy blood agar with 5% defibrinated sheep bl ood . Specimens were obtained with a rayon-tipp ed swab from the vaginal introitus or anus. After collection specimens were incu- bated in a selective broth m edium to facilitate rapid bacterial growth. Samples positive for group B strepto- cocci were identified within 12 to 24 hours by slide coagglutination. A se cond vaginal (n = 166) and rectal (n = 35) cul- ture for group B streptococci was obtain ed immediately after artificial or spontaneous rupture of membranes. None of the patients received antibiotics before culture . Statistical analysis was p erf ormed with X 2 and Fisher's exact test. Results During the study period, May to Augu st 1991, 1109 women had intrapartum vaginal screens for group B streptococci at our in stitution . Maternal colonization was documented in 305 of 1109 (27.5%) admission cu lt ures. Similarly, 49 of the 166 women (30%) in the study group had positive vagin al cultures for group B streptococci on admission. However, only 28 of the 49 cultures remained positive aft er rupture of membranes;