METRONIDAZOLE TREATMENT OF BACTERIAL VAGINOSlS IN PREGNANCY , AND PRETERM BIRTH : A RANDOMIZED , PLACEBO-CONTROLLED TRIAL

s From

CONCLUSIONS: PCR detects AF bacteria in 36% of patients with negative cultures and IL-6 levels 2000 pg/ml. Cytokine levels and pregnancy outcomes are similar in the +culture and -culture/+PCR groups. The association between AF infection and preterm labor may be underestimated by AF culture.
RESULTS: Twenty-one of 385 women (5.4%) were found to have ASB. CONCLUSION: ATP bioluminescence is an effective screening test for ASB in pregnancy. Implementation of ATP testing prior to culture, thereby selecting women with the greatest likelihood of ASB, would result in 50% reduction in urine cultures.
Future work will focus on optimizing differential cell lysis in order to improve specificities and further reduce the need for routine culture. Objectives" To determine whether metronidazole treatment of bacterial vaginosis(BV) in pregnancy reduces the rate of spontaneous preterm birth (PTB).

METRONIDAZOLE TREATMENT OF BACTERIAL
Study Design Randomised, double-blind, placebo-controlled study of two successive monthly of oral metronidazole (400 mg bd for days) in pregnant at 24,weeks' gestation, with intermediate flora (heavy growth of G. vaginalis) bacterial vaginosis (by Gram statn). Follow-up vaginal swabs were performed at 28, 32 and 36 weeks' gestation.
Exclusions included with age <17 years, multiple pregnancy, recent antibiotics for vaginal discharge Results 625 fully completed the study protocol at of the participating centres. 17(5 5%) of 312 in the placebo group gave birth preterm (<37 weeks), compared with 16(5 1%) of 313 in the metronidazole group A direct indicative of BV present in 341 women, however the PTB rate similar in both the placebo (10/165)and metronidazole (10/176) groups. In contrast, treatment with metronidazole in with obstetric/demographic risk factors for PTB, resulted in reduction of the PTB rate. Among with previous preterm dehvery (PPTD), 5(38.5%) of 13 given placebo gave birth preterm, compared with one(7 7%) of 13 given metronidazole Other categories showing reducuons in PTB rate "single" [10% placebo 6.2% metronidazole]" "age 20 years" 11.4% 0%]. and "unemployed" [9.7% 3.9%], although these differences not statistically significant Conclusions: In at low risk of PTB, treatment of BV in pregnancy appears to result in httle reducUon in the PTB rate, (although the power of this negative conclusion is limited due to sample size). In with risk factors (especially PPTD), treatment of BV orgarusms during pregnancy may reduce the PTB rate, however further enrolments necessary to demonstrate statistically significant reduction ,'ERVICAL ANTIBODIES TO HERPES SIMPLEX VIRUS DURING THE .I-IIRD TRIMESTER OF PREGNANCY. KA Boess MD, DH Watts MD, ZA trown MD, L Corey MD, and RL Ashley PhD University Of Washington ledical Center Depts. of Ob/Gyn & Laboratory Medicine Seattle, WA )BJECTIVE: To examine cervical secretions for anti-HSV IgA and determine elationship to HSV shedding during the third trimester of pregnancy. ;TUDY DESIGN: HSV-2 seropositive pregnant women had cervical ecretions collected weekly for anti-HSV IgA from 26-31 weeks' until lelivery. Patients collected daily genital specimens for HSV culture and PCR. rESULTS: Of 9 HSV seropositive women (8 HSV-2 only and HSV-1 and tSV-2), 5 (56%) women had a history of symptomatic herpes and 4 (44%) tad no history of genital symptoms. Cervical anti-HSV IgA was detected on. 0 (35.7%) of 84 days sampled. Anti-HSV IgA was detected on significantly nore days in women without a history of symptomatic genital herpes (21 of 7, 56.8%) than those with a history of genital HSV (9 of  Objective: Culturing mothers for Herpes Simalex virus (HSV) at the time of delivery identifies some, but not all of the infants at risk for developing neonatal herpes. We hypothesized that in addition to maternal cultures, culturing the neonate in the delivery room would enhance our ability to identify infants at risk for neonatal HSV.
Study Design: All obstetrical patients without signs or symptoms of herpes had cervical and vulvar cultures obtained on admission in labor. In addition, HSV cultures from the oropharynx of the newborn were obtained in the delivery room. HSV was isolated using diploid fibroblasts by standard technique. Results: Out of 9,667 total deliveries between 1989 and 1993, 6,338 women and 7,285 neonates had cultures obtained. Fifty-seven women had positive cultures, of which 7 had HSV isolated from the cervix, 44 from the vulva, and 6 from both sites. Over 90% of the "missed" maternal cultures were due to women who presented with vaginal bleeding, advanced cervical dilation, imminent delivery, clinical HSV symptoms, or admission for repeat cesarean section. Of the 9,667 deliveries, both maternal and infant cultures were obtained in 5,201. Of these patients, the virologic Lactobacillus concentrations were stable (+ log) over the cycle, facultative bacteria including Enterococcus, GBS, and E. coli, varied over the cycle and anaerobic bacteria were increased during menses. Women with bacterial vaginosis maintained high levels of anaerobic flora throughout the cycle, pH was stable in the absence of bleeding and correlated with Gram stain score. Myeloperoxidase data are incomplete, but levels are variable within each subject and have not correlated with cervical WBC count, pH, Gram stain score or specific flora. Three (43%) of 7 HIV positive women had HIV detected by culture of the cervix at least once, but only 4 (8.5%) of 48 cultures were positive. Four (57%) of 7 had cervical HIV detected by RNA PCR, and 25 (47%) of 53 specimens were positive by PCR. Most women were persistently po;itive or negative by PCR, but one subject had HIV detected intermittently. Ctr., Dallas, TX OBJECTIVE: 3TC (lamivudine) is currently being used as treatment for hepatitis B and in combination therapies for HIV. We sought to study the transfer of 3TC ((-)-2'-deoxy-3'-thiacytidine) across the human placenta both alone and in the presence of AZT (zidovudine).
STUDY DESIGN: Nine placentas from term, elective repeat cesarean deliveries were analyzed using the ex vivo single cotyledon perfusion system. Antipyrine was utilized as the reference compound for the determination of the clearance indices of 3TC alone and in combination with AZT. 3TC and AZT concentrations in the perfusates and tissues were quantified by reverse phase HPLC. RESULTS: The clearance index of 3TC at a maternal concentration of 1.39 /g/ml. was 0.23 _+ 0.14. At a peak concentration of 14.6/g/mL the CI was 0.14 + 0.06. When the maternal concentration of AZT was #g/mL, the CI of 3TC at the peak and trough concentrations was 0.18 _+ 0.08 and 0.19 _+ 0.10, respectively. When the concentration of AZ'I was 10/g/ml_ the CI of 3TC was 0.15 _+ 0.07 and 0.17 + 0.09, respectively, for the trough and peak concentrations. In addition, when the perfusion system was closed, there was little accumulation in the fetal perfusate, and in the intervillous space of the placental tissue. CONCLUSION: These data suggest that the CI of 3TC is not altered by the addition of AZT at its peak and trough concentrations. These data also suggest that there is little accumulation of 3TC in the fetal circulation and placental tissue.
Study Design: We studied ten women diagnosed with post partum endomyometritis. All women received 4.5 mg/kg actual body weight of gentamicin intravenously once daily as part of their antibiotic regimen. Serum peak and trough levels were obtained after the first two doses, as well as an eight hour level after the second dose. Serum creatinine levels were monitored, and ototoxicity was assessed clinically.
Results: The doses ranged from 255-600 mg (mean=360 rag), and the duration of therapy was 2-5 days. The mean elimination constant (Ke) was 0.105 (SD +/-0.008), and the volume of distribution (Vd) was 0.4 l/kg (SD +/-0.07). The mean peak levels after the first and second doses were 11.6 tg/ml (SD +/-2.3) and 13.0 lag/tnl (SD +/-2.5), respectively. Both trough levels were 0.3 ILtg/ml. The mean half-life was 6.6 hrs. There was no nephrotoxicity or ototoxicity observed. Clinical response was seen in 90% of patients; one patient responded with the addition of hcparin tbr presumed septic pelvic thrombophlebitis.

Conclusions:
Once daily dosing of gentamicin results in pharmacokinetic parameters similar to those observed in non-pregnant patients. No significant drug accumulation or toxicity occured. Therefore, single daily dose gentamicin in the post partum patient appears safe to use at 4.5 mg/kg actual body weight. Further studies necessary to confirm the safety and efficacy demonstrated in this report.  Results: The Dr+ E. cOli established long term (52 weeks) colonization of renal tissue. In the Drgroup, 50% of animals cleared infection within 20 weeks and 100% between 32 to 52 weeks, In the Dr + group, E. coli cells found to colonize the renal interstitium (detected by hematoxylin-eosin and immuno staining). Fimbrial antigen detected in the parenchymai regions affected by tubulointerstitial nephritis but not in the non-affected tissue. Histological changes in the Dr + group included interstitial inflammation, interstitial fibrosis, and tubular atrophy in the kidney tissue. In the Dr group, histological lesions significantly less than those of the Dr group and involved smaller number of ammais.

Conclusions: The obtained results
consistent with the hypothesis that mutation within the dra region, coding for Dr fimbriae, factor that mediates binding to renal interstitium, prevents the development of the renal changes characteristically in tubulointerstitial nephritis. It remains to be investigated whether the gestational pyelonephritis may contribute to long term complications including hypertension and end stage renal disease. Supported by NIH Grant R01 DK42029 (BN). Objectives: To study relationships between microbial flora of fetal membranes, histopathologic findings of the membranselplacentalumbilical cord, and clinical symptoms among women delivering at term. Study Design: We studied consecutive women with singleton pregnancy delivering at term during a 3-month period. Immediately after delivery of the placenta, the nurse-midwife separated the fetal membranes. Specimens for culture were collected between the membranes, close to the placental edge with cotton-tipped wooden swabs treated with charcoal. The specimens were kept at room temperature in Stuart's medium and transported to the laboratory within 24 hours (within 48 hours during week-ends). Formalin-fixed membrane roles, and placental and umbilical tissue specimens were studied by a pathologist who was blinded as to clinical and microbiologic findings.
Results: We obtained culture specimens from the fetal membranes of 312 (96.9%) of the 330 eligible women who delivered during the study period. At least one bacterial species was recovered in 218 cases (69.9%). Pathogenic microorganisms were recovered from 106 (34%) of the 312 membranes. Culture specimens from the membranes of eight women delivered by elective cesarean section showed no growth. Sixty-two (35.8%) of the first 173 consecutively delivered women had histopathologic evidence of chorioamnionitis. We recovered pathogens from the fetal membranes of 35 (37.2%) women with and 21 (30%) without evidence of chorioamnionitis in the membranes (O.R.= 1.4, p<0.4). Women with fever during labor (n=7) were significantly more likely than those not having fever to have chorioamnionitis (100% vs. 32.1%, O.R.= 16.0). Obleca=: To axss the effi3, and safety of mnpleillin/sulbaatm (A/S) zltpared to ampicillhl/getllattli (A/O) In the tt,eatmtmt of horioanmionith. Study design: Patients with chorioamnionitis (temprzature > 100'F on two ooesions, at least one hour apmt or single temperature 8rearer than 101F and ruptured membranes and at least one ofthe following; maternal leukoeytosis, mmemal tachycardia, fetal tahyoat'die, foul smelling anmloti fluid) wore randomly assigned in double blind fashion t receive either A/S (n 121) A/G (n 120). Clinical failures were defined as those subjects with persistent temperature elevation (> 100.4"F) for more than 24 hours postpaxtum. Adverse went= tn both subject and neonate we assessed.
Patients undeqinl; sereau section (8/53) were more likely to fail therapy than those delivering vaglnally (0/102)[p < 0.001]), more likely to develop postpartum endnmyometrltis; (5/53 v 0/102 [p 0.004]), mad have a longm" hospital duration (5.2 + 0.7 v,s 3.4 + 0.6 [p < 0.001]), All clinical failures occta'md ha subjects undergoing e,e,sareml section. Endomyometritls oc,-tm-txl in 3/27 eesareen patieu receiving A/S atd in 2/26 patients t'teiving A/G (p NS). lqo n'iou= =tdweme evente due to the study drugs were observed. Minor adver=e events were equally distributed n the two study groups, lqo neonates died from sepsis. Objectives: We studied the behavior of vaginal douching to determine: 1) the prevalence, methods and rationale for this behavior, and 2) whether douching is associated with an increased prevalence of vaginal bacteriosis. Study Design: This is a cross-sectional study of women attending outpatient gynecologic clinics at the University of Chicago. Structured interviews were administered by trained female interviewers. Vaginal smears were collected, batched, gram stained, and analyzed by Nugent's criteria for vaginal bacteriosis. Results: Of the 103 women interviewed, 78.6% had douched in their lifetime, and 74.2% of these women douched at least once a month. Compared with women who did not douche regularly, women who douched at least once month were more likely to use genital deodorant products (OR=36.6, 95% CI 4.7-286.4, p<0.001), to report the belief that douching can eliminate odors (OR=3.5, 95% CI 1.2-9.9, p=0.015), and to report the belief that douching can prevent vaginal infections (OR=5.3, 95% CI 1.8-16.0, p=0.002). Women who had douched during the past month were more likely to have vaginal bacteriosis than all other women in the study (OR=4.1, 95%CI 1.0, 16.6, p=0.041). Conclusions: Beliefs about genital hygiene appear likely to underlie most douching: these include deodorizing and preventing infection. We found a significant association between recent douching and vaginal bacteriosis. Any public health efforts designed to discourage douching will have to consider the complex interplay of cultural and racial differences in health beliefs and douching behaviors. OBJECTIVE: In this report we compared clinical manifestations and tubal abnormalities observed at laparoscopy among women with acute salpingitis. STUDY DESIGN: We compared the type, and when applicable, the severity of clinical manifestations in 82 women with definite laparoscopic evidence of acute salpingitis. Women with findings other than salpingitis (n=22), normal findings (n=9) or possible, but not definite salpingitis (n=42) were excluded from these analyses.
RESULTS: Two general categorie.s of laparoscopic findings were present: 1) tubal occlusion and moderate to severe adhesions tended to occur together, in 30 patients and 2) pelvic/abdominal exudate tended to occur separately, in 27 patients. Tubal occlusion was positively associated with older age, palpable adnexal mass and negatively associated with rebound tenderness, the tenderness score and the isolation of Neisseriae gonorrhoeae and Chlamydia trachomatis. Moderate to severe adhesions were positively associated with duration of abdominal pain and negatively associated with the tenderness score. Exudate in the pelvis/abdomen was positively associated with the tenderness score, the white blood count and N.
gonorrhoeae and negatively associated with the duration of pain, oral contraceptive use and palpable adnexal mass. Reanalyses among patients with no prior history of PID did not change the findings. CONCLUSION: Clinical manifestation traditionally used to judge the clinical severity of PID partially predict the laparoscopic findings, tend to distinguish those with tubal occlusion and/or moderate to severe adhesion from those with peritonitis, and provide insight into the pathophysiology. However, the predictive value of clinical manifestations are low and not reliable for the individual patient. Gonococcal PID is most serious and costly complication of gonorrhea. Pathogenesis of PID is poorly understood. Two major factors contribute to the virulence of gonococci, 1) resistance to the bactericidal effect of normal human serum and 2) attachment to human tissues. Clq-mediated virulence is the first significant difference found between the PID and local strains (Perkin Elmer Award to S. Nowicki's laboratory for the best research in STD in 1995). The sac-4 DNA region of N. gonorrhoeae is known to confer partial serum resistance (serR). We have recently identified 344 base pair segment in the 3' end of the sac-4 region which upon transformation with plasmid pRP350 conferred Clq dependent virulence to laboratory strain F62 Recently we found two unique local isolates of N. gonorrhoeae which carried 344 bp fragment but were avirulent in our pup model. We hypothesize that the 344 bp fragment from local strains carry point mutation(s) resulting in abolishing Clq dependent virulence.
Study design: N. gonorrhoeae strains 1655 and 1653 were used to PCR amplify 344 bp fragments. PCR products were subcloned to appropriate vector. Resultant p.lasmids pRPI659, pRPI665 were sequenced, and transformed to laboratory strata F62. Virulence of these constructs was examined on pup model. Resulls: A 344 bp fragments from local isolates did not confer Clq mediated virulence to N. gonorrhoeae F62 upon transformation. In control plasmid pRP350 from PID isolate transformed strain F62 to Clq dependent virulent strain. Sequencing of plasmids pRP1655 and pRPI659 from local strains and pRP350 from PID strain has been carried out and two deletions and three substitutions were identified. Deletion of the 5' segment of pRP350 that overlap identified mutations abolished Clq dependent OBJECTIVE Our aim was to study the association between severity of pelvic inflammatory disease (PID) at laparoscopy and the length of time to live birth. STUDY DESIGN: Beginning in 1960, cohort of 1,730 women in Lund, Sweden who had clinical symptoms of acute PID and who desired pregnancy was followed tbr up to 24 years All study participants underwent laparoscopy and were assigned to of four categories normal findings (n 442), or mild (n 371), moderate (n 580), or severe (n 337) tubal inflammation Cumulative live birth rates were obtained by life-table analysis, and proportional hazards ratios were compared among women with varying degrees of tubal inflammation. RESULTS. Median time to achieve live birth according to severity of tubal inflammation at laparoscopic evaluation of PID was 4.0 years for women with no tubal inflammation, 4 years for women with mild, 6 years tbr women with moderate, and 11.8 years for women with severe inflammation Three years after the acute episode of PID, 37% with no inflammation, 38% with mild, 26% with moderate, and 16% with severe tubal inflammation were successful in achieving live birth Cumulative proportion of women achieving live birth after 12 or more years was 90%, 90%, 82%, and 56% respectively, tbr those with no, mild, moderate, and severe inflammation After adjustment for age and predisposing factors, women with severe disease and subsequent episodes of PID were eight times more likely to remain childless as those experiencing single episode with mild tubal inflammation CONCLUSIONS Increasing severity of PID correlates with longer time in achieving live birth These data underscore the importance of PID prevention and Objectives' The purpose of this study was to determine whether the risk of preterm delivery associated with drug abuse during pregnancy could be accounted for by charaeterstics of drug abusing women such as ethnicity, high number of pregnancies, increased tobacco use, and high frequency of genital infections. Study Design: Two cohorts of women were compared for their frequency of preterm delivery: (I) 181 women seeking treatment for drug abuse (1990)(1991)(1992)(1993)(1994) and (2) 557 non-drug abusing women (1984)(1985)(1986). Both cohorts had evaluations of genital specimens performed in one laboratory at UWMC. Women participated in personal interviews and had genital specimens obtained between 16 and 30 weeks gestational age. At delivery, a structured medi6al record review was conducted and preterm delivery was defined as gestational age <37 weeks and birth weight <2500 gm. Results: Drug abusing women were at higher risk of preterm delivery (Unadjusted Odds Ratio (OR) 3.2, 95% Confidence Interval (CI) 1.7 6.2) Drug abusing women were also more likely to have the following characteristics: older age, nonwhite race, high number of pregnancies, tobacco use, 60% increase in STD's, and 2.6 fold increase in bacterial vaginosis After multivariable adjustment for competing risks of preterm delivery, drug abusing women were still at increased risk of preterm delivery (OR 2.6; 95% CI I. 6.8) When drug abuse (OR 4.2; 95% CI 5-12) was modelled together with bacterial vaginosis (OR 3.0; 95% CI 1.0-93), it remained an independent risk for preterm delivery. The risk of preterm delivery among women with both factors was no greater than the separate risks. Conclusions: The risk of preterm delivery associated with drug abuse during pregnancy is independent from characteristics of drug abusing women: bacterial vaginosis, STD's, and demographic and social factors. correlate with antibody to the 60 kD chlamydial heat shock protein (Chsp-60) in women at risk for chlamydial (CT) cervicitis and acute PID. STUDY DESIGN: Cross-sectional study of 156 randomly selected STD clinic attendees without clinical evidence of PID and 150 women with PID confirmed by laparoscopy (n=69) or endometrial biopsy (n=81). All women underwent a standard interview, cultures and determinations of antibody to CT using microimmunoflorescence and of antibody to Chsp-60 using ELISA. A positive Chsp-60 level was an optical density (OD) _> 0.2. RESULTS: Chsp-60 antibody was present in 146 (47%) of the patients. In multivariate analyses, Chsp-60 antibody was related to confirmed PID (OR=2.3, CI 1.3-2.4), age > 20 (OR=2.1, CI 1.1-3.8), CT IgG titers of 1:16-1:128 (OR=2.6, CI 1.2-5.7), CT IgG titers > 1:128 (OR=44.7 CI 8.6-233), > I0 lifetime sexual parters, non-white race and current OC use, but not to CT culture or IgM titers to CT. Chsp-60 antibody was also related to laparoscopic findings among those with salpingitis confirmed by laparoscopy.

PREVALENCE AND CORRELATES
Chsp <0.2 Chsp 0.2-1.0 m=croorgan=sms (except Lactobaci//us spp.) isolated from the genital tract at inclusion, during 2nd and 3rd limester and at birth likely to be among the highly sexually active participants in comparison to the remaining participants (p 0.05). CONCLUSIONS: 1) Sexual activity decreases 3-fold during pregnancy among with the most frequent coitus. 2) There is strong association between higher sexual activity and prevalences of various microorganisms isolated from the genital tract and (3)  Study Design: There is ongoing randomized clinical trial. All attending their first and 36 week prenatal visit had cervical swabs taken and analyzed with the Gen-Probe PACE (Gen-Probe, Inc., San Diego, CA) DNA probe for Chlamydia trachomatis. All patients with positive results were asked to participate. Patients assigned to receive either one gram oral dose of azithromycin powder seven days of erythromycin base, 500 mg q.i.d. Patients unable to tolerate original randomization medications due to gastrointestinal side effects allowed to crossover to the opposite study medication. Test of cure swabs were obtained two weeks after completion of treatment. Patients testing positive for Chlamydia trachomatis at follow-up considered treatment failures and retreated with the other agent. The incidence of side effects and compliance rates assessed by patient questionnaire at the two week follow-up visit. Objectives: The purpose of the study was to document the impact of the introduction into clinical practice of the proposed CDC guidelines for prevention of early onset GBS disease. Study Design: A retrospective study comparing the prevalence of early onset GBS sepsis (positive blood culture GBS) at Magee-Womens Hospital prior to the institution of CDC guidelines (January 1, 1992-June 30, 1995) with the prevalence of GBS sepsis following introduction of the CDC protocol (October 1, 1995-March 31, 1996. The microbiology laboratory data base was reviewed for all positive GBS blood cultures from the nursery, as were the medical records of blood culture GBS positive newborns. Results: From January 1, 1992-June 30, 1995 there were 36 cases of early onset GBS sepsis among 31,133 births for a rate of 1.15 per 1000 live births. After introduction of the CDC protocol there was case of early onset GBS sepsis in 5500 births or. 18 per 1000 live births (p=0.03). During these time periods the prevalence of maternal GBS vaginal colonization remained the same (27% vs 25%). By March, 1996 intrapartum prophylaxis for GBS was the most common new antibiotic order accounting for 36% of new antibiotic orders at MWH. Post introduction of the CDC protocol, IV penicillin made up 83.2% of these orders, clindamycin 13.2%, erythromycin 0.5%, amyicillin 1.3% and other 1.8%. Conclusions: The proposed CDC guidelines for prevention of early onset GBS infection in newborns is very effective, reducing early onset GBS 5-fold. Institution of the CDC protocol was easily accomplished in large community hospital where two-thirds of deliveries are by private attendings. DOES   Objectives: The purpose of the study was to evaluate maternal determinants of infant Group B Streptococci colonization at delivery. Study Design" A cohort of 536 mother/infant pairs were enrolled at Magee-Womens iHospital from July 1995 to February 1996. Infants had specimens taken before 12 hr .of life for the isolation of GBS from the outer ear, throat, anus, and umbilicus. Specimens were grown on solid media following broth enrichment. Medical records were reviewed to determine the mother's antepartum (AP) GBS culture and intrapartum (IP) chemoprophylaxis status. A subgroup of 216 mothers also had IP vaginal specimens taken for GBS isolation when they were admitted for delivery. Results: Maternal GBS colonization rates in the AP and IP were 26% and 24%, respectively. Seven percent (38/536) of infants were GBS positive from at least one site. Infants were more likely to be positive if maternal AP cultures had been positive (12% versus 5%; P=0.02) but 56% of positive infants were born to culture negative mothers. Infants treated with IP antimicrobials for > 4 hr had 54% reduction in colonization (95% CI 0.2-1.3) compared to infants whose mothers were untreated or treated <4 hr AP cultures were 77% sensitive and 82% specific for predicting women positive at delivery but 23% of women who were positive at delivery were negative earlier in pregnancy. Women who were positive at any testing time were more likely to have culture positive infant when compared to those who were negative at both testing times (19% versus 0.7%; P < 0.0001) Conclusions: Antepartum maternal specimens taken for the isolation of GBS are inadequate predictors of which women will be positive at delivery. Negative antepartum maternal culture sta,'us may result in false reassurance that the infant is not at risk for invasive GBS disease. The mechanism of enhancement of acquisition of HIV infection by coinfection with sexually transmitted diseases (STDs) is unknown. SLPI, serine protease inhibitor, is known to protect oral mucosal surfaces from transmission of the AIDS virus through its inhibition of the protease activity required for HIV infection of mucosal monocytes and macrophages. However, little is known about SLPI function in the genital tract. Objective: This study was undertaken to determine if the abundant, secreted cysteine proteases of T. vaginalis degrade purified SLPI or SLPI found in vaginal secretions and render it non-functional. Study Design: Proteases were induced from isolates of T. vaginalis collected from pregnant patients. Protozoan-free, supernatant proteases were fractionated by isoelectric point over pH gradient of 3.0 to 9.0 using BioRad RotopHer. Dialyzed, pH gradient fractions of proteases were mixed with human recombinant SLPI (2 ug) with freshly collected vaginal secretions from normal women (n=3) and incubated at 37C for 20 hours. The degradation patterns of the mixtures were followed by SDS-PAGE and Western immunoblot analysis using SLPI sp.ecific polyclonal antiserum. Protease inhibitory function of SLPI was tested using serine protease peptide-specific substrates in the presence or absence of Trichomonas proteases and excess E-64 inhibitor. Results: Two T. vaginalis proteases (Mr of 32 and 100kDa) with isoelectric points of approximately pH 7.0, completely degraded the purified SLPI Western signal within 20 hours. These proteases were also able to degrade endogenous SLPI in vaginal secretions. Cysteine protease inhibitor, E-64, completely protected recombinant SLPI from degradation or loss of functional activity. Conclusions: These studies demonstrate that T. vaginalis cysteine proteases can degrade SLPI in vitro, and could play role in the in vivo degradation of SLPI, thereby rendering the vaginal immune cells more susceptible to HIV infection. These findings could explain the increased risk of HIV acquisition seen in association with T. vaginalis infection and could provide mechanism for enhancement of HIV infection by other non-ulcerative STDs. Results: After 96 hours of incubation HIV replication increased 25 fold with PBMC coincubation (p<0.01), 48 fold with PMN coincubation (p<0.01), 27 fold with CT extract and PBMC coincubation (p<0.Ol), and 33 fold with CT extract and PMN coincubation. Conclusion: Inflammatory cells found in the female genital tract, in association with CT infection, significantly enhance HIV replication in vitro. CT alone in combination with inflammatory cells did not further enhance HIV replication. This data suggests that the inflammatory response to CT infection is part responsible for Objectives: In some reports, HIV-infected women have an increased incidence of recurrent vaginal infections. The purpose of this retrospective review was to identify the rate of vaginal infection and treatment response in a cohort ofHIV-infected women in Houston, Texas.

SECRETORY LEUKOCYTE PROTEASE INHIBITOR (SLPI) IN VAGINAL SECRETIONS IS DEGRADED BY CYSTEINE
Study Design: We reviewed medical records of 96 HIV-infected women who had 149 visits from July 1994 to March 1996. We recorded rate and type of vaginal infections, the therapeutic regimens and immune status of all patients.
Results: The mean (+ S.D.) age for the entire group was 26.3 +/-7.7 years, and the ethnic composition was African-American 81.3%, Hispanic 10.4% and Caucasian 8.3%. There were 52 patients with 59 episodes ofvaginal infection.We found bacterial vaginosis in 35, Candidiasis in 15 and Trichomonis in 9. The overall failure rate was of 59 (5.1%). The failures all occurred in patients who received oral therapy (2 with B.V. and with Candida). One therapeutic failure had a CD4 count of> 500, the other two had CD4 counts < 200. In this study there was no association between the occurrence of vaginal infections and immune status (NS). In addition, there were no associations between CD4 counts and treatment failure (NS).
Conclusion: This preliminary review demonstrates a 5.1% treatment failure rate for vaginal infections in HIV-positive women. We found no significant association between treatment failures, therapeutic regimens or immune status. examined in term pregnancies. STUDY DESIGN: 33 women undergoing cesarean section had specimens of AF, chorioamnion (CA), decidua (D) and placenta (Plac) collected at delivery. AF, CA and Plac were cultured. AF TNF-a and IL-6 concentrations were determined by ELISA. Immunohistologic differentiation was performed on all tissues for Tand Blymphocytes and macrophages using antibody markers and/or granulocytes (PMN) identified by morphology. TNF-a ,d IL-6 producing cells were identified by mRNA in situ hybridization using S labeled probes. Fetal macrophages were identified by combined in situ immunohistology and hybridization using macrophage specific antibody HAM56 and a DNA probe for Y-chromatin. RESULTS: Both AF infection and labor were associated with elevated AF levels of TNF-a and IL-6. Macrophages and PMN's were the predominant cell types in the CA. IL-6 mRNA signal in the decidua was related to labor while IL-6 mRNA signal within macrophages in the amnion and chorion was related to positive AF and CA culture. Macrophages in the amnion were related to positive AF culture. Using the double stain technique, Y-chromatin positive macrophages were present in all 19 pregnancies with male fetus and in none of 14 pregnancies with female fetus (p<0.001). CONCLUSIONS: Fetal amnion and chorion macrophages synthesize CK in response to infection. Most of these macrophages in the amnion and chorion appear to be of fetal origin. However, there is little consensus about screening during pregnancy, and the tests used to establish fetal diagnosis of toxoplasmosis complex and time consuming. Prenatal diagnosis of congenital toxoplasmosis is based ultrasonography, fetal blood sampling or inoculation of cultured cells or mice with amniotic fluid. Given the risk associated with fetal-blood sampling and the delay in obtaining definitive results with conventional parasitologic methods ultrasound, better methods needed.

DETECTION OF TOXOPLASMA GONDH IN AMNIOTIC FLUID BY
T. gondii has been diagnosed by PCR but detection of the amplified product by an ELISA has, to our knowledge, not been reported. ELISA detection has increased sensitivity other methods.
Study Design: Using human amniotic fluids that had been spiked with T. gondii, we established parameters for the amplification and detection of fragment of the 35-fold repetitive B gene.The 115-base pair product, digoxigenin-labeled, was hybridized to biotinylatcd internal probe and detected by ELISA. The hybridization validated the specificity of the amplified PCR product. The possible presence of PCR inhibitors in individual amniotic fluids was evaluated by amplifying a portion of the human beta actin gene.
Results: '/ gondii DNA readily detected in amniotic fluids by our PCR/ELISA. PCR analysis of DNA extracted from various bacteria, yeast, and viruses using the T. gondii primer pairs did not result in amplification of any crossreacting DNA.
Similarly, amniotic fluids obtained for genetic analysis were uniformly negative for 7: gondii.
Conclusions: The PCR/ELISA will allow for the definitive diagnosis of congenital T.
gondii infection in day. This sviil reduce the incidence of unnecessary pregnancy terminations based uncertainty of fetal infection as well as allow the prompt initiation of antibiotic treatment to infected fetuses. The risk of transmission of hepatitis B virus (HBV) from amniocentesis in HBV carriers has not been thoroughly investigated. In report from Taiwan, the risk of perinatal HBV transmission in HBV carriers not higher after amniocentesis compared to control group of HBV carriers. Immtmoprophylaxis failures occured in 10% of infants in both groups. Our objective to the risk of transmission of HBV in chronic carriers who undergo anmiocenteses at hospital.

Study Design
This study prospective, longitudinal study, and data collected about who HBV carriers and underwent amniocentesis. The infants of these followed in special clinic from birth to year of age. Maternal data examined included HBV antigen and antibody status, liver function tests (LFTs), and the amniocentesis procedure report.
Pediatric data obtained from clinic records including the neonatal and 12 month HBsAg and vaccine record.

Results 25
identified. Two of 25 neonates stillborn unrelated to hepatitis, and infants lost to follow-up, leaving 18 mother-child pairs to evaluate. All 18 chronic HBV carriers at the time of anmiocentesis and delivery. No mother had abnormal LFTs, and only of 18 positive for HBeAg. 10 amniocentesis for advanced maternal age, and for abnormal MSAFP screening. None of the amniocenteses recorded bloody, and the placenta anterior in of 18 procedures. None of the 18 infants (95% CI:O-18.5%)were positive for HbsAg during the first month of life at 12 months of age. All infants received HBV vaccine and HBIg immunoprophylaxis.

Conclusion
The risk of transmission of HBV to the fetus after amniocentesis in who HBV carriers is low. Immunoprophylaxis in these infants successful. Objectives: In the rabbit model of infection-induced preterm labor, as well as in women with labor associated with inflammation and infection, the expression of pro-inflammatory cytokines such as tumor necrosis factor-alpha (TNF-0 and interleukin-1 (IL-1) stimulate prostaglandin synthesis leading to uterine contractions. On the other hand, endogenous anti-inflammatory proteins such as transforming growth factor-beta (TGF-) and interleukin-1 receptor antagonist (IL-lra)inhibit prostaglandin production. We hypothesize that uterine contractility depends upon delicate balance between stimulatory and inhibitory cytokine-like factors in the uterine environment; the purpose of this study was to compare the concentration of TNF-C to the anti-inflammatory factor TGFin amniotic fluid using the rabbit model. Study Design: New Zealand white rabbits at approximately 70% gestation underwent hysteroscopic inoculation of E. coli and were sacrificed at varying times after bacterial introduction. Amniotic fluid was collected and assayed for protein levels of TNF-O and TGFby enzyme-linked immunoassay (ELISA, R&D Systems, Minneapolis, MN). Results: TNF-C rose from 0.44 pg/ml at time 0 to 2.5 pg/ml by 16 hrs (P_qg.019). TGFwas significantly higher than TNF-C at time 0 (610 pg/ml versus 0.44 pg/ml, P.f:0.01), and did not change over time. Conclusions: (1) TNF-cconcentration, as measured by ELISA, rises significantly with infection. (2) Compared to TNF-TGFis present at high levels in the amniotic fluid but does not change with infection. We speculate that rising TNF-c in the face of unchanging TGFcreates an imbalance between proand antiinflammatory cytokines which leads to uterine contractions. Objective: to study the relationship between serological response to chlamydia160 kDa heat shock protein (hsp 60) and tubal factor infertility (TFI).

AMNIOTIC FLUID CONCENTRATIONS OF TNF-a VERSUS TGF-[ IN
Study Design: Twenty three women with TFI and 33 women with male factor infertility (controls) were studied. TFI was defined as unprotected intercourse for one year with hydrosalpinx or distal tubal occlusion on hysterosalpingogram or at laparoscopy. Demographic data was gathered by chart review. Patients' sera were tested for antibodies to the hsp 60 using an enzym.e linked immunoabsorbent assay (ELISA). Stepwise logistic regression was performed on patient's age, race/ethnicity, self reported history'ofChlamydia or pelvic inflammatory disease, history ofectopic pregnancy and status of anti-hsp antibodies with the hsp 60 ELISA.  (FF) were obtained at the time of oocyte aspiration from 149 women undergoing IVF. FF free of visible blood contamination. Samples assayed for IgG and IgA antibodies to C. trachomatis by two different ELISA assays employing either recombinant Chlamydia-specific lipopolysaccharide (LPS) fragment the chlamydial major outer membrane protein (MOMP). All subjects were tested for cervical C. trachomatis by DNA probe; FF were tested for C_..tr.achomatis by polymerase chain reaction (PCR) and ligase chain reaction (LCR). Results: Sera from 90 (60%) subjects positive for antichlamydial IgG utilizing the LPS assay; 54 (36%) positive for antichlamydial lgG in the MOMP assay.
Similarly, LPS-directed and MOMP-dirccted IgA antibodies in detected in 50 (34%) and 21 (14%) of the subjects, respectively In FF, IgG antibodies to MOMP were also more prevalent than IgA antiMOMP antibodies (46% vs. 9%). However, unlike in sera IgA antiLPS antibodies present in FF from 107 (72%) vomen opposed to only 50 (34%) being IgG antiLPS positive (P<.0001). In 54 women antichlamydial IgA was detectable in FF but not in the corresponding serum. All cervical and FF samples negative for C. trachomatis by PCR and LCR. ..Conclusions: Similarities in the prevalence of IgG and IgA antibodies to MOMP in sera and FF suggest that these antibodies enter the FF primarily by transudation from the circulation. The greatly increased prevalence of lgA antibodies to chlamydial LPS in FF suggests that localized humoral immune immune response to this antigen may be induced in the upper genital tract in the absence of systemic tmmune response.  Study Design: Vaginal samples were obtained from 39 reproductiveaged women (1 Native American, 5 Asians, 6 Blacks, and 27 Caucasians). The selective Rogosa agar was used to isolate lactobacilli, from which phages were induced by mitomycin C. Phage virulence was analyzed by cross-infecting these phages against a range of vaginal lactobacilli.
Results: Of 19 samples from women with vaginal infections, 12 did not have lactobacilli. From the remaining 7 infection samples and the 20 samples from healthy women, 40 Lactobaci/lus stains were isolated, from which 6 Lactobacil/us species were identified and 7 phages were isolated. One phage, kc039, was characterized as follows: plaque morphology, small and clear, 1-2 mm in diameter; burst size, 300/cell; spontaneous induction rate, 10"e/cell; DNA, double stranded and linear, 41 kb; and shape, a hexagonal head and a noncontractile tail. This phage attacked in vitro 8 out of 39 vaginal Lactobaci//us strains from other women. One strain, kc013, was completely lysed by kc039. Conclu..s.i..ons: Using this bioassay the MIC's for fluconazole, amphotencin B and miconazole were determined and found to be within the pubhshed NCCLS reference ranges. One T. glabrata wild type was resistant to fluconazole and miconazole but not to amphotedcin B. We are currently increasing the number of isolates tested and correlating these in vitro results to clinical outcome. Tiballi  OBJECTIVE: Retained placenta is a common intrapartum complication following vaginal delivery which requires manual extraction (ME). Because of concerns related to bacterial contamination and subsequent infection, many clinicians administer antibiotic prophylaxis with this procedure. There are no data to support the need for antibiotic prophylaxis with ME. STUDY DESIGN: This retrospective study examined 190 patients who underwent ME following vaginal delivery. Seventy one patients received prophylactic antibiotics and 119 patients did not. Indications for ME were prolonged third stage (123), retained products (16), bleeding (14), avulsed cord (14), and other (23). None of these patients received intrapartum antibiotics or were suspected of infection during their labor. Indications for ME, risk factors for infection and concomitant curettage were similar for both study groups. The decision to administer antibiotic prophylaxis was based on individual clinician preference. If antibiotics were given, they were administered from the time of ME up to 48 hours post procedure. Patients receiving antibiotic prophylaxis were compared to those not receiving prophylaxis for post partum infectious morbidity.
RESULTS: Three patients in the prophylaxis group developed post partum endometritis (4.2%). Of these three patients, one also had uterine curettage. One subject was admitted on post partum day 10 for suspected endometritis. All three had 24 hours of prophylactic antibiotics. No patient in the group not receiving antibiotics developed post partum endometritis. CONCLUSION: These data suggest that ME following vaginal delivery does not place the patient at an increased risk for post partum endometritis. Therefore, prophylactic antibiotics are not necessary. Study Desigtt A prospective cohort study of 109 patients referred by their gynecologists for evaluation of chronic vaginal symptoms. Patients were interviewed about their use of OTCs and AMs over the preceding year, the amount of money spent on each, and whether their physicians had been informed of these treatments.
RESULTS: A positive response was significantly more likely if there was no prior history of syphilis or if there was high initial RPR titer (> 32). Only 33/54 (61%) observations at three months or greater had positive response.
CONCLUSION: Our study suggests that an absence of history of syphilis and an initial high RPR titer are predictive of positive response following recommended treatment. We speculate that we may be under-treating our pregnant patients infected with syphilis. Further studies are needed to evaluate the normal serologic response following syphilis treatment in pregnancy.